CN-121982206-A - Automatic monitoring system and method for tunnel surrounding rock deformation measurement

CN121982206ACN 121982206 ACN121982206 ACN 121982206ACN-121982206-A

Abstract

The application discloses an automatic monitoring system and method for tunnel surrounding rock deformation measurement, and belongs to the technical field of tunnel construction safety monitoring. The monitoring method comprises the steps of projecting an infrared structured light pattern to a monitoring section, fusing natural texture information of the surrounding rock surface to construct a virtual target and obtaining a reference image, arranging a reference terminal, an optical reference point and a reference terminal in a tunnel stable area to establish a dynamic space measurement reference, obtaining a real-time measurement image of the monitoring section by using the monitoring terminal, preprocessing, calculating sub-pixel image plane displacement by using a digital image correlation algorithm fused with natural texture analysis, carrying out dynamic error compensation according to pose change quantity fed back by the reference terminal, and finally outputting actual displacement. The application realizes the full section monitoring of the tunnel surrounding rock monitoring section without a physical target, and the sub-millimeter measurement accuracy, and has the advantages of strong environment interference resistance and all-weather automatic monitoring.

Inventors

MA JUNWEI
YIN DENGPING
ZHANG BEI
ZHANG GANG
XIA QINGSHUI
WANG YUAN

Assignees

西安悦创地理信息工程有限公司

Dates

Publication Date: 20260505
Application Date: 20260123

Claims (9)

1. The automatic monitoring method for tunnel surrounding rock deformation measurement is characterized by comprising the following steps: S1, projecting a preset infrared structure light pattern to a monitoring section of surrounding rock of a region to be monitored to form an infrared structure light band covering the monitoring section, constructing a virtual target of the monitoring section based on the infrared structure light band information of the infrared structure light band and an initial image of surrounding rock surface natural texture information of the monitoring section, fusing the infrared structure light band information and the natural texture information based on the initial image, and storing the virtual target as a reference image, wherein the initial image comprises the infrared structure light band information of the infrared structure light band and the initial image of surrounding rock surface natural texture information of the monitoring section; S2, at least one fixed optical datum point and a datum terminal for observing the optical datum point are arranged in a stable area where a tunnel is finished, infrared structure light is emitted to the section of the stable area through the optical datum point to form an optical datum, a unified measurement coordinate system is established based on the datum terminal, and the datum terminal and the monitoring terminal are synchronously collected to respectively acquire a real-time datum image of the optical datum and a real-time measurement image of a virtual target; S3, preprocessing the real-time measurement image acquired in the step S2 to acquire an infrared structure light characteristic line image with high resolution; s4, matching the characteristic line image obtained in the S3 with a reference image, and obtaining sub-pixel image plane displacement of each characteristic point in the real-time measurement image relative to the reference image by adopting a digital image correlation algorithm based on natural texture analysis; s5, according to the position change of each characteristic point in the real-time reference image, calculating the pose change quantity of the monitoring terminal, carrying out geometric correction on the displacement of the sub-pixel image surface in S4 based on the pose change quantity, and converting the corrected displacement of the sub-pixel image surface into actual displacement and outputting the actual displacement by combining the calibration parameters and the measurement distance of the monitoring terminal; s6, analyzing the time sequence of the actual displacement output by the S5 to obtain the deformation development trend of the monitoring section, and judging and early warning based on a preset early warning rule.
2. The automated tunnel surrounding rock deformation measurement and monitoring method according to claim 1, wherein in S1, the infrared structured light pattern is a grid-like pattern, and the infrared structured light pattern covers the dome, the dome shoulder and the sidewall areas of the monitoring section.
3. The automated tunnel surrounding rock deformation measurement monitoring method according to claim 1, wherein in S1, the infrared structured light is near infrared invisible laser light, and the wavelength of the near infrared invisible laser light is any one of 850nm, 910nm, and 915 nm; the laser linewidth of the infrared structure light band is less than or equal to 1.5nm, and the single laser linewidth on the projection surface is 0.8 mm-1.2 mm; The diameter of the light spot at the intersection point of the infrared structure light band is less than or equal to 2mm.
4. The method for automatically monitoring deformation of tunnel surrounding rock according to claim 1, wherein in S3, the preprocessing of the real-time measurement image includes performing image processing of the real-time measurement image based on a deep neural network, dividing a light band region of the infrared structural light band from a complex background of the real-time measurement image, and performing super-resolution reconstruction of the divided light band region to obtain a high-resolution infrared structural light characteristic line image.
5. The automated tunnel surrounding rock deformation measurement monitoring method according to claim 4, wherein the deep neural network is PSPNet network structure, and the image segmentation and super-resolution reconstruction process is as follows: firstly, inputting a real-time measurement image into PSPNet network structures, and obtaining a feature map after convolution downsampling; then, inputting the feature map into a pyramid pooling module, and adding the output of the pyramid pooling module with the original feature map; And finally, up-sampling by a convolution layer and bilinear interpolation, optimizing a segmentation result by using a Dice Loss function, and finally outputting a binarized infrared structured light characteristic line image.
6. The method for automatically monitoring deformation of surrounding rock of tunnel according to claim 1, wherein in S4, the calculation process of the sub-pixel image plane displacement is: S401, matching the characteristic line image with the reference image, describing deformation of a characteristic point image subarea by adopting a shape function, and obtaining an integral pixel displacement initial value of a characteristic point through an integral pixel searching algorithm; s402, establishing a quality evaluation model of natural textures of the image based on the feature points so as to evaluate the similarity degree and the image quality of the image subareas; s403, sub-pixel optimization is carried out on the actual displacement initial value of the whole pixel based on the Newton-Laportson iteration method, and sub-pixel image plane displacement with sub-pixel precision is obtained.
7. The automated tunnel surrounding rock deformation measurement monitoring method according to claim 6, wherein in the quality evaluation model in S402, the similarity degree of the characteristic point natural texture image before and after deformation is evaluated, and a zero-mean normalized least square distance correlation function C ZNSSD is adopted; and (5) evaluating the quality of the natural texture image of the characteristic points, and adopting the average gray gradient of the image subareas.
8. The automatic monitoring method for tunnel surrounding rock deformation measurement according to claim 1, wherein in S5, the pose change amount is data after identifying and rejecting abnormal values by 3 sigma criterion; The specific process of converting the corrected sub-pixel image plane displacement into the actual displacement comprises the following steps: firstly, calculating a proportionality coefficient K of sub-pixel surface displacement and actual displacement; Finally, the sub-pixel image plane displacement is converted into actual displacement by using a proportionality coefficient K, wherein the actual displacement comprises vault subsidence displacement and peripheral convergence displacement.
9. The automatic monitoring system for measuring the deformation of the surrounding rock of the tunnel is characterized by being used for realizing the automatic monitoring method for measuring the deformation of the surrounding rock of the tunnel according to any one of claims 1-8, and comprises the following components: The virtual target and optical reference construction unit comprises at least one virtual target construction unit for projecting infrared structure light patterns to the monitoring section of the tunnel to form a virtual target, and an optical reference point for projecting infrared structure light to the section of the stable region where the tunnel is finished to form an optical reference; the virtual target construction unit is arranged on the side wall of the area to be monitored, and the optical datum point is arranged on the side wall of the finished stable area; The system comprises an acquisition unit, a detection unit and a control unit, wherein the acquisition unit comprises at least one monitoring terminal and at least one reference terminal, the monitoring terminal and the reference terminal are respectively provided with an optical filter which only transmits an infrared structure optical band, the monitoring terminal is arranged in a region to be monitored of a tunnel and is used for acquiring real-time measurement images of a virtual target on a monitoring section, and the reference terminal is arranged in a stable region of the tunnel which is finished and is used for acquiring real-time reference images of an optical reference point; A data processing and control unit respectively connected to the monitoring terminal, the reference terminal, the target construction unit, and the optical reference point, the data processing and control unit configured to: the virtual target construction unit and the optical reference point are controlled to form a virtual target in the monitoring area and an optical reference in the stabilizing area respectively; Preprocessing a real-time measurement image, executing a digital image correlation algorithm based on natural texture analysis to calculate the sub-pixel image plane displacement of each characteristic point in the real-time measurement image relative to a reference image, geometrically correcting the sub-pixel image plane displacement, converting the corrected sub-pixel image plane displacement into actual displacement and outputting the actual displacement; The analysis and early warning module is connected with the data processing and control unit, and is used for obtaining the deformation development trend of the monitoring section based on time sequence analysis of the converted actual displacement and judging and early warning based on preset early warning rules.

Description

Automatic monitoring system and method for tunnel surrounding rock deformation measurement Technical Field The application relates to an automatic monitoring system and method for deformation measurement of tunnel surrounding rock, and belongs to the technical field of deformation monitoring of tunnel surrounding rock. Background In the tunnel construction process, high-precision and real-time monitoring of surrounding rock deformation at an initial support stage is a key for guaranteeing construction safety. In the prior art, the mainstream automatic monitoring technology mainly comprises a total station automatic monitoring method, a three-dimensional laser scanning method and a traditional binocular vision measurement method. The total station automatic monitoring method is characterized in that the characteristic points are discrete points, so that a monitoring blind area exists, and continuous deformation of the whole section cannot be reflected. The three-dimensional laser scanning method performs deformation analysis by acquiring high-density point cloud data of the tunnel surface, and can realize surface measurement, but the laser signal of the method is seriously attenuated in tunnel dust and water vapor environments, so that long-term stable monitoring precision is difficult to ensure. The traditional binocular vision measurement method simulates human eyes through two cameras, and performs three-dimensional reconstruction and deformation measurement by utilizing the parallax principle, and has the defects that when illumination in a tunnel is uneven and textures are absent, characteristic points are difficult to match, so that measurement stability and reliability are insufficient. The technology relies on an entity target or specific environmental conditions, is greatly interfered by ambient light and dust, and can not meet the requirements of intelligent tunnel construction on measurement continuity and precision. Disclosure of Invention In order to solve the problems in the prior art, the application provides an automatic tunnel surrounding rock deformation measurement monitoring system and method which do not depend on a physical target, have strong environmental interference resistance and can realize full-section high-precision continuous monitoring. According to one aspect of the application, there is provided an automated monitoring method of deformation of surrounding rock of a tunnel, the monitoring method comprising: The monitoring method comprises the following steps: S1, projecting a preset infrared structure light pattern to a monitoring section of surrounding rock of a region to be monitored to form an infrared structure light band covering the monitoring section, constructing a virtual target of the monitoring section based on the infrared structure light band information of the infrared structure light band and an initial image of surrounding rock surface natural texture information of the monitoring section, fusing the infrared structure light band information and the natural texture information based on the initial image, and storing the virtual target as a reference image, wherein the initial image comprises the infrared structure light band information of the infrared structure light band and the initial image of surrounding rock surface natural texture information of the monitoring section; S2, at least one fixed optical datum point and a datum terminal for observing the optical datum point are arranged in a stable area where a tunnel is finished, infrared structure light is emitted to the section of the stable area through the optical datum point to form an optical datum, a unified measurement coordinate system is established based on the datum terminal, and the datum terminal and the monitoring terminal are synchronously collected to respectively acquire a real-time datum image of the optical datum and a real-time measurement image of a virtual target; S3, preprocessing the real-time measurement image acquired in the step S2 to acquire an infrared structure light characteristic line image with high resolution; s4, matching the characteristic line image obtained in the S3 with a reference image, and obtaining sub-pixel image plane displacement of each characteristic point in the real-time measurement image relative to the reference image by adopting a digital image correlation algorithm based on natural texture analysis; s5, according to the position change of each characteristic point in the real-time reference image, calculating the pose change quantity of the monitoring terminal, carrying out geometric correction on the displacement of the sub-pixel image surface in S4 based on the pose change quantity, and converting the corrected displacement of the sub-pixel image surface into actual displacement and outputting the actual displacement by combining the calibration parameters and the measurement distance of the monitoring terminal; s6, analyzing the time sequence of the