Method and device for evaluating surface quality of initial shotcrete during underground gate excavation

Abstract

The present invention discloses a method and device for evaluating the surface quality of initial sprayed concrete in underground gate excavation operations based on a depth camera; wherein the method comprises: performing mechanism modeling processing on the depth camera based on the principle of the depth camera to obtain the mechanism model of the depth camera; performing camera calibration processing on the depth camera based on the mechanism model of the depth camera to obtain the calibrated depth camera; performing image data acquisition on the surface to be detected based on the calibrated depth camera to obtain the image data of the surface to be detected; calculating the reflectivity at different pixel points in the image data of the surface to be detected based on the mechanism model of the depth camera; performing initial sprayed surface quality evaluation processing on the image data of the surface to be detected based on the reflectivity to obtain an evaluation result. In the embodiment of the present invention, it is achieved that the quality of the initial spraying construction is not affected by adverse light environment factors such as dark light or uneven illumination under underground construction conditions, which may lead to erroneous judgments, and can effectively ensure the quality of the initial spraying construction.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a method and device for evaluating the surface quality of initial shotcrete in underground gate excavation operations based on a depth camera. Background Art

[0002] In underground projects, especially during the blasting and excavation of valve rooms and tunnels, initial jetting support concrete (initial jetting) is used as an initial support process to provide basic guarantee for the construction quality of the entire underground excavation project. Therefore, the quality assessment of the initial jetting surface is critical to ensuring construction safety and quality; the current commonly used initial jetting quality assessment method relies on the experience and knowledge of experts, manually completes data collection through the use of mechanical measuring tools, and completes subjective analysis and evaluation based on the collected data. It faces problems such as low detection efficiency, high labor costs, and the use of subjective judgments limited by technical level and experience and knowledge, which makes it difficult to meet the needs of modern intelligent construction operations; with the development of current construction technology, the construction process is gradually developing towards automation and digitization, and the thickness and surface flatness of the initial jetting support concrete have high requirements, so it is necessary to establish an evaluation method and system that can automatically detect the quality of the initial jetting surface. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art. The present invention provides a method and device for evaluating the surface quality of initial sprayed concrete in underground gate excavation operations based on a depth camera. The initial sprayed surface information is collected by a depth camera, and a quality evaluation conclusion is obtained through analysis and processing, which provides a reference basis for construction quality management and ensures the quality of the initial sprayed support concrete process. It is not affected by adverse light environment factors such as dark light or uneven lighting under underground construction conditions, which may lead to erroneous judgments, and can effectively ensure the quality of the initial spraying construction.

[0004] In order to solve the above technical problems, an embodiment of the present invention provides a method for evaluating the surface quality of initial shotcrete in underground gate excavation operations based on a depth camera, the method comprising:

[0005] Based on the principle of the depth camera, a mechanism modeling process is performed on the depth camera to obtain a mechanism model of the depth camera;

[0006] Performing camera calibration processing on the depth camera based on the mechanism model of the depth camera to obtain a calibrated depth camera;

[0007] The image data of the surface to be inspected is collected based on the calibrated depth camera to obtain the image data of the surface to be inspected;

[0008] Calculating and obtaining the reflectivity of different pixels in the image data of the surface to be detected based on the mechanism model of the depth camera;

[0009] The initial spraying surface quality evaluation processing is performed on the image data of the surface to be detected based on the reflectivity to obtain an evaluation result.

[0010] Optionally, the mechanism model of the depth camera is as follows:

[0011]

[0012] Among them, R(θ i ) indicates that the incident angle at pixel i is θ i The reflectivity at P i Represents the detection intensity result of the depth camera; λ i (d i ,θ i ) represents the correlation between detection intensity, detection distance and incident angle; V decay Represents the attenuation coefficient of pixel i.

[0013] Optionally, the incident angle is θ i hour:

[0014]

[0015] in, Represents the viewing angle vector in the depth camera coordinate system, which is the vector from the depth camera position to the pixel i; Represents the normal vector at pixel i.

[0016] Optionally, based on Lambertian scattering and specular reflection, a reflection attenuation model at different angles is constructed to describe V decay ,as follows:

[0017]

[0018] Where a is a constant, which is set according to the detection environment. In the environment of underground engineering with poor natural light, it is set to 2; n is the scattering coefficient, but when n = 1, it is regarded as a Lambertian scattering, and when n → + ∞, it is a specular reflection; that is, set:

[0019] n=1 / cos(θ i ),in

[0020] Optionally, the mechanism model of the depth camera is rewritten as follows:

[0021]

[0022] Among them, λ i (d i ,θ i) is obtained by numerical analysis based on the collected data, as follows:

[0023]

[0024] Among them, A i , B i , C i , D i , E i , F i are the internal parameters of the depth camera, which together constitute the internal parameter matrix of the depth camera.

[0025] Optionally, the performing camera calibration processing on the depth camera based on the mechanism model of the depth camera to obtain a calibrated depth camera includes:

[0026] The depth camera collects a set of image data at a preset distance within the range of the calibration plane, and at the same time, the depth camera collects six sets of image data at intervals of 5 degrees from the normal line angle in the preset angle interval to obtain calibration collected image data;

[0027] Input the calibration collected image data into the mechanism model of the depth camera for solving, and average each value to obtain an internal parameter matrix;

[0028] A calibrated depth camera is obtained based on the internal parameter matrix.

[0029] Optionally, the calibration collected image data is input into the mechanism model of the depth camera for solution, and the average of each value is calculated to obtain an internal parameter matrix, as follows:

[0030]

[0031] After the calibration acquisition image data is input into the formula for solution, the average value of each value is obtained to obtain the internal parameter matrix as follows:

[0032]

[0033] Among them, λ i (d i ,θ i ) represents the correlation between detection intensity, detection distance and incident angle; P i Represents the detection intensity test result of the depth camera; M_CAM represents the internal parameter matrix; Indicates A i , B i , C i , D i , E i , F i The corresponding average value.

[0034] Optionally, the calculating based on the mechanism model of the depth camera to obtain the reflectivity at different pixel points in the image data of the surface to be detected includes:

[0035] Preprocessing the image data of the surface to be detected by removing points outside the detection distance to obtain the preprocessed image data of the surface to be detected;

[0036] The preprocessed image data of the surface to be detected is input into the mechanism model of the depth camera, and the reflectivity at different pixel points in the preprocessed image data of the surface to be detected is obtained.

[0037] Optionally, the initial spraying surface quality assessment process of the image data of the surface to be detected based on the reflectivity is as follows:

[0038]

[0039]

[0040] Where S represents the standard deviation of the initial spray surface; R(θ i ) indicates that the incident angle at pixel i is θ i The reflectivity at ; M represents the mean; sum(i) represents the number of detected pixels;

[0041] According to the requirements of the construction industry, a threshold value T0 is set. When the standard deviation of the initial sprayed surface is less than the set threshold value, S≤T0, the initial sprayed surface quality in the underground excavation operation meets the requirements, otherwise it does not meet the requirements.

[0042] In addition, an embodiment of the present invention further provides a device for evaluating the surface quality of initial shotcrete in underground gate excavation operations based on a depth camera, the device comprising:

[0043] Modeling module: used for performing mechanism modeling processing on the depth camera based on the principle of the depth camera to obtain a mechanism model of the depth camera;

[0044] Calibration module: used for performing camera calibration processing on the depth camera based on the mechanism model of the depth camera to obtain a calibrated depth camera;

[0045] Acquisition module: used to acquire image data of the surface to be detected based on the calibrated depth camera to obtain image data of the surface to be detected;

[0046] Calculation module: used for calculating and obtaining the reflectivity of different pixel points in the image data of the surface to be detected based on the mechanism model of the depth camera;

[0047] Evaluation module: used for performing initial spraying surface quality evaluation processing on the image data of the surface to be detected based on the reflectivity to obtain an evaluation result.

[0048] In an embodiment of the present invention, the initial spraying surface information is collected by a depth camera, and a quality assessment conclusion is obtained through analysis and processing, which provides a reference basis for construction quality management and ensures the quality of the initial spraying support concrete process. Compared with the traditional manual inspection method, it has higher efficiency and is convenient for automatic archiving of data, which is conducive to the digitization and intelligence of construction projects. It is not affected by adverse light environment factors such as dark light or uneven lighting under underground construction conditions, which may lead to erroneous judgments. It can effectively ensure the quality of the initial spraying construction and also promote the construction safety of underground projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 is a flow chart of a method for evaluating the surface quality of initial shotcrete in underground gate excavation operations based on a depth camera in an embodiment of the present invention;

[0051] Figure 2 It is a schematic diagram of the structural composition of a device for evaluating the surface quality of initial shotcrete in underground gate excavation operations based on a depth camera in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] Embodiment 1

[0054] See also Figure 1 , Figure 1 It is a flow chart of a method for evaluating the surface quality of initial shotcrete in underground gate excavation operations based on a depth camera in an embodiment of the present invention.

[0055] like Figure 1 As shown, a method for evaluating the surface quality of initial shotcrete in underground gate excavation operations based on a depth camera comprises:

[0056] S11: performing mechanism modeling processing on the depth camera based on the principle of the depth camera to obtain a mechanism model of the depth camera;

[0057] In the specific implementation of the present invention, the mechanism model of the depth camera is as follows:

[0058]

[0059] Among them, R(θ i ) indicates that the incident angle at pixel i is θ i The reflectivity at P i Represents the detection intensity result of the depth camera; λ i (d i ,θ i ) represents the correlation between detection intensity, detection distance and incident angle; V decay Represents the attenuation coefficient of pixel i.

[0060] Furthermore, the incident angle is θ i hour:

[0061]

[0062] in, Represents the viewing angle vector in the depth camera coordinate system, which is the vector from the depth camera position to the pixel i; Represents the normal vector at pixel i.

[0063] Furthermore, in the attenuation coefficient part, based on Lambertian scattering and specular reflection, a reflection attenuation model at different angles is constructed to describe V decay ,as follows:

[0064]

[0065] Where a is a constant, which is set according to the detection environment. In the environment of underground engineering with poor natural light, it is set to 2; n is the scattering coefficient, but when n = 1, it is regarded as a Lambertian scattering, and when n → + ∞, it is a specular reflection; that is, set:

[0066] n=1 / cos(θ i ),in

[0067] Furthermore, the mechanism model of the depth camera is rewritten as follows:

[0068]

[0069] Among them, λ i (d i ,θ i) is obtained by numerical analysis based on the collected data, as follows:

[0070]

[0071] Among them, A i , B i , C i , D i , E i , F i are the internal parameters of the depth camera, which together constitute the internal parameter matrix of the depth camera.

[0072] Specifically, the present invention adopts a commercial depth camera based on the time-of-flight (ToF) principle. The principle of this camera is to adopt active light detection. The illumination unit of the TOF camera emits an intensity modulated signal to illuminate the entire scene, and then uses the cumulative delay value between the incident light signal and the reflected light signal to measure the distance.

[0073] According to the principle of depth camera, the detection value of the depth camera is defined as the detection intensity. It is feasible to judge the reflectivity of the detected surface by the detection intensity, because at each pixel point i of the depth camera detection result, the detection intensity is proportional to the signal amplitude received after scattering from the detected surface, and is affected by the surface reflectivity at the corresponding viewing angle, the intensity distribution of the lighting signal, the distance, and the detection efficiency of the camera. Based on this principle, the following model is constructed to model the surface reflectivity:

[0074]

[0075] Among them, R(θ i ) indicates that the incident angle at pixel i is θ i The reflectivity at P i Represents the detection intensity result of the depth camera; λ i (d i ,θ i ) represents the correlation between detection intensity, detection distance and incident angle; V decay Represents the attenuation coefficient of pixel i.

[0076] The incident angle is θ i hour:

[0077]

[0078] in, Represents the viewing angle vector in the depth camera coordinate system, which is the vector from the depth camera position to the pixel i; Represents the normal vector at pixel i.

[0079] For the attenuation coefficient, based on Lambertian scattering and specular reflection, a reflection attenuation model at different angles is constructed to describe V decay ,as follows:

[0080]

[0081] Where a is a constant, which is set according to the detection environment. In the environment of underground engineering with poor natural light, it is set to 2; n is the scattering coefficient, but when n = 1, it is regarded as a Lambertian scattering, and when n → + ∞, it is a specular reflection; that is, set:

[0082] n=1 / cos(θ i ),in

[0083] Then rewrite the mechanism model of the depth camera as follows:

[0084]

[0085] Among them, λ i (d i ,θ i ) is obtained by numerical analysis based on the collected data, as follows:

[0086]

[0087] Among them, A i , B i , C i , D i , E i , F i are the internal parameters of the depth camera, which together constitute the internal parameter matrix of the depth camera.

[0088] S12: performing camera calibration processing on the depth camera based on the mechanism model of the depth camera to obtain a calibrated depth camera;

[0089] In the specific implementation process of the present invention, the depth camera is calibrated based on the mechanism model of the depth camera to obtain a calibrated depth camera, including: the depth camera collects a set of image data at a preset distance within the range of the calibration plane, and the depth camera collects six sets of image data from the preset angle interval at intervals of 5 degrees to obtain calibrated collected image data; the calibrated collected image data is input into the mechanism model of the depth camera for solution, and the average value of each value is obtained to obtain an internal parameter matrix; the calibrated depth camera is obtained based on the internal parameter matrix.

[0090] Furthermore, the calibration collected image data is input into the mechanism model of the depth camera for solution, and the average of each value is calculated to obtain the internal parameter matrix, as follows:

[0091]

[0092] After the calibration acquisition image data is input into the formula for solution, the average value of each value is obtained to obtain the internal parameter matrix as follows:

[0093]

[0094] Among them, λ i (d i ,θ i ) represents the correlation between detection intensity, detection distance and incident angle; P i Represents the detection intensity test result of the depth camera; M_CAM represents the internal parameter matrix; Indicates A i , B i , C i , D i , E i , F i The corresponding average value.

[0095] Specifically, generally, manufacturers have unified standards and designs for camera internal parameters. However, due to errors in the processing process, there are inevitably slight differences in parameters. Therefore, it is necessary to ensure that the same camera is used during data acquisition, and data collected by different cameras cannot be mixed together for data analysis. At the same time, a prerequisite is assumed, that is, the internal parameters of the same camera do not change during a single acquisition. The camera calibration process is as follows:

[0096] Place the camera at the distance calibration plane [d min ,d max ] (for ease of explanation, assume 0.5 to 1.0), collect a set of data every 10cm, and at the same time, the camera is at an angle of 0-30° from the normal line of the plate, with an interval of 5 degrees, and collect 6 sets of data on each side, for a total of 6x(6+6) 72 sets of data (if necessary, the data can be improved by adjusting the collection distance and angle to obtain more accurate results).

[0097]

[0098] After the calibration acquisition image data is input into the formula for solution, the average value of each value is obtained to obtain the internal parameter matrix as follows:

[0099]

[0100] Among them, λ i(d i ,θ i ) represents the correlation between detection intensity, detection distance and incident angle; P i Represents the detection intensity test result of the depth camera; M_CAM represents the internal parameter matrix; Indicates A i , B i , C i , D i , E i , F i The corresponding average value.

[0101] S13: collecting image data of the surface to be detected based on the calibrated depth camera to obtain image data of the surface to be detected;

[0102] In the specific implementation process of the present invention, during the detection process, the operator carries a calibrated depth camera to collect data on the surface to be detected. The data collection requirements are: 1) Focus on the distance detection distance [D min ,D max ] range; 2) The detection angle of view covers [0,180] as much as possible. It may be adjusted according to the situation of the depth camera and the working scene; the image data of the surface to be detected can be obtained.

[0103] S14: Calculating and obtaining the reflectivity of different pixels in the image data of the surface to be detected based on the mechanism model of the depth camera;

[0104] In the specific implementation process of the present invention, the reflectivity at different pixel points in the image data of the surface to be detected is calculated based on the mechanism model of the depth camera, including: preprocessing the image data of the surface to be detected to remove points outside the detection distance, and obtaining the preprocessed image data of the surface to be detected; inputting the preprocessed image data of the surface to be detected into the mechanism model of the depth camera, and obtaining the reflectivity at different pixel points in the preprocessed image data of the surface to be detected.

[0105] Specifically, after acquiring the image data of the surface to be detected, the image data of the surface to be detected is first preprocessed, that is, the points outside the detection distance are eliminated because the points outside the detection distance will cause greater interference; then the reflectivity of different pixel points in the preprocessed image of the surface to be detected is calculated through the formula of the mechanism model of the depth camera.

[0106] S15: Performing initial spraying surface quality assessment processing on the image data of the surface to be inspected based on the reflectivity to obtain an assessment result.

[0107] In the specific implementation process of the present invention, the initial spraying surface quality assessment process of the image data of the surface to be detected based on the reflectivity is as follows:

[0108]

[0109] in

[0110] Where S represents the standard deviation of the initial spray surface; R(θ i ) indicates that the incident angle at pixel i is θ i M represents the mean value; sum(i) represents the number of detected pixels; according to the requirements of the construction industry, a threshold value T0 is set. When the standard deviation of the initial sprayed surface is less than the set threshold value, S≤T0, the initial sprayed surface quality in underground excavation operations meets the requirements, otherwise it does not meet the requirements.

[0111] Specifically, the reflectivity is numerically analyzed to evaluate the roughness of the detection surface. When the reflectivity is consistent, it means that the surface is smooth. Therefore, after obtaining the reflectivity data of the surface to be detected through the depth camera, the mean and variance of the reflectivity of all detection results are calculated. The mean and variance are general data processing methods:

[0112]

[0113] Where, S represents the standard deviation of the initial spray surface; R(θ i ) indicates that the incident angle at pixel i is θ i M represents the mean value; sum(i) represents the number of detected pixels.

[0114] According to the requirements of the construction industry, a threshold value T0 is set. When the standard deviation of the initial sprayed surface is less than the set threshold value, S≤T0, the initial sprayed surface quality in the underground excavation operation meets the requirements, otherwise it does not meet the requirements.

[0115] In an embodiment of the present invention, the initial spraying surface information is collected by a depth camera, and a quality assessment conclusion is obtained through analysis and processing, which provides a reference basis for construction quality management and ensures the quality of the initial spraying support concrete process. Compared with the traditional manual inspection method, it has higher efficiency and is convenient for automatic archiving of data, which is conducive to the digitization and intelligence of construction projects. It is not affected by adverse light environment factors such as dark light or uneven lighting under underground construction conditions, which may lead to erroneous judgments. It can effectively ensure the quality of the initial spraying construction and also promote the construction safety of underground projects.

[0116] Embodiment 2

[0117] See also Figure 2 , Figure 2 It is a schematic diagram of the structural composition of a device for evaluating the surface quality of initial shotcrete in underground gate excavation operations based on a depth camera in an embodiment of the present invention.

[0118] like Figure 2 As shown, a device for evaluating the surface quality of initial shotcrete in underground gate excavation operations based on a depth camera, the device comprising:

[0119] Modeling module 21: used to perform mechanism modeling processing on the depth camera based on the principle of the depth camera to obtain a mechanism model of the depth camera;

[0120] In the specific implementation of the present invention, the mechanism model of the depth camera is as follows:

[0121]

[0122] Among them, R(θ i ) indicates that the incident angle at pixel i is θ i The reflectivity at P i Represents the detection intensity result of the depth camera; λ i (d i ,θ i ) represents the correlation between detection intensity, detection distance and incident angle; V decay Represents the attenuation coefficient of pixel i.

[0123] Furthermore, the incident angle is θ i hour:

[0124]

[0125] in, Represents the viewing angle vector in the depth camera coordinate system, which is the vector from the depth camera position to the pixel i; Represents the normal vector at pixel i.

[0126] Furthermore, in the attenuation coefficient part, based on Lambertian scattering and specular reflection, a reflection attenuation model at different angles is constructed to describe V decay ,as follows:

[0127]

[0128] Where a is a constant, which is set according to the detection environment. In the environment of underground engineering with poor natural light, it is set to 2; n is the scattering coefficient, but when n = 1, it is regarded as a Lambertian scattering, and when n → + ∞, it is a specular reflection; that is, set:

[0129] n=1 / cos(θ i ),in

[0130] Furthermore, the mechanism model of the depth camera is rewritten as follows:

[0131]

[0132] Among them, λ i (d i ,θ i ) is obtained by numerical analysis based on the collected data, as follows:

[0133]

[0134] Among them, A i , B i , C i , D i , E i , F i are the internal parameters of the depth camera, which together constitute the internal parameter matrix of the depth camera.

[0135] Specifically, the present invention adopts a commercial depth camera based on the time-of-flight (ToF) principle. The principle of this camera is to adopt active light detection. The illumination unit of the TOF camera emits an intensity modulated signal to illuminate the entire scene, and then uses the cumulative delay value between the incident light signal and the reflected light signal to measure the distance.

[0136] According to the principle of depth camera, the detection value of the depth camera is defined as the detection intensity. It is feasible to judge the reflectivity of the detected surface by the detection intensity, because at each pixel point i of the depth camera detection result, the detection intensity is proportional to the signal amplitude received after scattering from the detected surface, and is affected by the surface reflectivity at the corresponding viewing angle, the intensity distribution of the lighting signal, the distance, and the detection efficiency of the camera. Based on this principle, the following model is constructed to model the surface reflectivity:

[0137]

[0138] Among them, R(θ i ) indicates that the incident angle at pixel i is θ i The reflectivity at P i Represents the detection intensity result of the depth camera; λ i (d i ,θ i ) represents the correlation between detection intensity, detection distance and incident angle; V decay Represents the attenuation coefficient of pixel i.

[0139] The incident angle is θ i hour:

[0140]

[0141] in, Represents the viewing angle vector in the depth camera coordinate system, which is the vector from the depth camera position to the pixel i; Represents the normal vector at pixel i.

[0142] For the attenuation coefficient, based on Lambertian scattering and specular reflection, a reflection attenuation model at different angles is constructed to describe V decay ,as follows:

[0143]

[0144] Where a is a constant, which is set according to the detection environment. In the environment of underground engineering with poor natural light, it is set to 2; n is the scattering coefficient, but when n = 1, it is regarded as a Lambertian scattering, and when n → + ∞, it is a specular reflection; that is, set:

[0145] n=1 / cos(θ i ),in

[0146] Then rewrite the mechanism model of the depth camera as follows:

[0147]

[0148] Among them, λ i (d i ,θ i ) is obtained by numerical analysis based on the collected data, as follows:

[0149]

[0150] Among them, A i , B i , C i , D i , E i , F i are the internal parameters of the depth camera, which together constitute the internal parameter matrix of the depth camera.

[0151] Calibration module 22: used for performing camera calibration processing on the depth camera based on the mechanism model of the depth camera to obtain a calibrated depth camera;

[0152] In the specific implementation process of the present invention, the depth camera is calibrated based on the mechanism model of the depth camera to obtain a calibrated depth camera, including: the depth camera collects a set of image data at a preset distance within the range of the calibration plane, and the depth camera collects six sets of image data from the preset angle interval at intervals of 5 degrees to obtain calibrated collected image data; the calibrated collected image data is input into the mechanism model of the depth camera for solution, and the average value of each value is obtained to obtain an internal parameter matrix; the calibrated depth camera is obtained based on the internal parameter matrix.

[0153] Furthermore, the calibration collected image data is input into the mechanism model of the depth camera for solution, and the average of each value is calculated to obtain the internal parameter matrix, as follows:

[0154]

[0155] After the calibration acquisition image data is input into the formula for solution, the average value of each value is obtained to obtain the internal parameter matrix as follows:

[0156]

[0157] Among them, λ i (d i ,θ i ) represents the correlation between detection intensity, detection distance and incident angle; P i Represents the detection intensity test result of the depth camera; M_CAM represents the internal parameter matrix; Indicates A i , B i , C i , D i , E i , F i The corresponding average value.

[0158] Specifically, generally, manufacturers have unified standards and designs for camera internal parameters. However, due to errors in the processing process, there are inevitably slight differences in parameters. Therefore, it is necessary to ensure that the same camera is used during data acquisition, and data collected by different cameras cannot be mixed together for data analysis. At the same time, a prerequisite is assumed, that is, the internal parameters of the same camera do not change during a single acquisition. The camera calibration process is as follows:

[0159] Place the camera at the distance calibration plane [d min ,d max] (for ease of explanation, assume 0.5 to 1.0), collect a set of data every 10cm, and at the same time, the camera is at an angle of 0-30° from the normal line of the plate, with an interval of 5 degrees, and collect 6 sets of data on each side, for a total of 6x(6+6) 72 sets of data (if necessary, the data can be improved by adjusting the collection distance and angle to obtain more accurate results).

[0160]

[0161] After the calibration acquisition image data is input into the formula for solution, the average value of each value is obtained to obtain the internal parameter matrix as follows:

[0162]

[0163] Among them, λ i (d i ,θ i ) represents the correlation between detection intensity, detection distance and incident angle; P i Represents the detection intensity test result of the depth camera; M_CAM represents the internal parameter matrix; Indicates A i , B i , C i , D i , E i , F i The corresponding average value.

[0164] Acquisition module 23: used to acquire image data of the surface to be detected based on the calibrated depth camera to obtain image data of the surface to be detected;

[0165] In the specific implementation process of the present invention, during the detection process, the operator carries a calibrated depth camera to collect data on the surface to be detected. The data collection requirements are: 1) Focus on the distance detection distance [D min ,D max ] range; 2) The detection angle of view covers [0,180] as much as possible. It may be adjusted according to the situation of the depth camera and the working scene; the image data of the surface to be detected can be obtained.

[0166] Calculation module 24: used for calculating and obtaining the reflectivity of different pixel points in the image data of the surface to be detected based on the mechanism model of the depth camera;

[0167] In the specific implementation process of the present invention, the reflectivity at different pixel points in the image data of the surface to be detected is calculated based on the mechanism model of the depth camera, including: preprocessing the image data of the surface to be detected to remove points outside the detection distance, and obtaining the preprocessed image data of the surface to be detected; inputting the preprocessed image data of the surface to be detected into the mechanism model of the depth camera, and obtaining the reflectivity at different pixel points in the preprocessed image data of the surface to be detected.

[0168] Specifically, after acquiring the image data of the surface to be detected, the image data of the surface to be detected is first preprocessed, that is, the points outside the detection distance are eliminated because the points outside the detection distance will cause greater interference; then the reflectivity of different pixel points in the preprocessed image of the surface to be detected is calculated through the formula of the mechanism model of the depth camera.

[0169] Evaluation module 25: used for performing initial spraying surface quality evaluation processing on the image data of the surface to be detected based on the reflectivity to obtain an evaluation result.

[0170] In the specific implementation process of the present invention, the initial spraying surface quality assessment process of the image data of the surface to be detected based on the reflectivity is as follows:

[0171]

[0172] Where, S represents the standard deviation of the initial spray surface; R(θ i ) indicates that the incident angle at pixel i is θ i M represents the mean value; sum(i) represents the number of detected pixels; according to the requirements of the construction industry, a threshold value T0 is set. When the standard deviation of the initial sprayed surface is less than the set threshold value, S≤T0, the initial sprayed surface quality in underground excavation operations meets the requirements, otherwise it does not meet the requirements.

[0173] Specifically, the reflectivity is numerically analyzed to evaluate the roughness of the detection surface. When the reflectivity is consistent, it means that the surface is smooth. Therefore, after obtaining the reflectivity data of the surface to be detected through the depth camera, the mean and variance of the reflectivity of all detection results are calculated. The mean and variance are general data processing methods:

[0174]

[0175] Where, S represents the standard deviation of the initial spray surface; R(θ i ) indicates that the incident angle at pixel i is θ i M represents the mean value; sum(i) represents the number of detected pixels.

[0176] According to the requirements of the construction industry, a threshold value T0 is set. When the standard deviation of the initial sprayed surface is less than the set threshold value, S≤T0, the initial sprayed surface quality in the underground excavation operation meets the requirements, otherwise it does not meet the requirements.

[0177] In an embodiment of the present invention, the initial spraying surface information is collected by a depth camera, and a quality assessment conclusion is obtained through analysis and processing, which provides a reference basis for construction quality management and ensures the quality of the initial spraying support concrete process. Compared with the traditional manual inspection method, it has higher efficiency and is convenient for automatic archiving of data, which is conducive to the digitization and intelligence of construction projects. It is not affected by adverse light environment factors such as dark light or uneven lighting under underground construction conditions, which may lead to erroneous judgments. It can effectively ensure the quality of the initial spraying construction and also promote the construction safety of underground projects.

[0178] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0179] In addition, the above is a detailed introduction to a method and device for evaluating the surface quality of initial sprayed concrete in underground gate excavation operations based on a depth camera provided by an embodiment of the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for evaluating the surface quality of initial shotcrete in underground gate excavation operations based on a depth camera, characterized in that: The method comprises: Based on the principle of the depth camera, a mechanism modeling process is performed on the depth camera to obtain a mechanism model of the depth camera; Performing camera calibration processing on the depth camera based on the mechanism model of the depth camera to obtain a calibrated depth camera; The image data of the surface to be inspected is collected based on the calibrated depth camera to obtain the image data of the surface to be inspected; Calculating and obtaining the reflectivity of different pixels in the image data of the surface to be detected based on the mechanism model of the depth camera; Performing initial spraying surface quality assessment processing on the image data of the surface to be inspected based on the reflectivity to obtain an assessment result; The mechanism model of the depth camera is as follows: Among them, R(θ i ) indicates that the incident angle at pixel i is θ i The reflectivity at P i Represents the detection intensity result of the depth camera; λ i (d i ,θ i ) represents the correlation between detection intensity, detection distance and incident angle; V decay Represents the attenuation coefficient of pixel i; The performing camera calibration processing on the depth camera based on the mechanism model of the depth camera to obtain a calibrated depth camera includes: The depth camera collects a set of image data at a preset distance within the range of the calibration plane, and at the same time, the depth camera collects six sets of image data at intervals of 5 degrees from the normal line angle in the preset angle interval to obtain calibration collected image data; Input the calibration collected image data into the mechanism model of the depth camera for solving, and average each value to obtain an internal parameter matrix; Obtaining a calibrated depth camera based on the internal parameter matrix; The calibration acquisition image data is input into the mechanism model of the depth camera for solution, and the average value of each value is calculated to obtain the internal parameter matrix, as follows: After the calibration acquisition image data is input into the formula for solution, the average value of each value is obtained to obtain the internal parameter matrix as follows: Among them, λ i (d i ,θ i ) represents the correlation between detection intensity, detection distance and incident angle; P i Represents the detection intensity test result of the depth camera; M_CAM represents the internal parameter matrix; Indicates A i , B i , C i , D i , E i , F i The corresponding average value.

2. The method for evaluating the surface quality of initial shotcrete during underground gate excavation according to claim 1 is characterized in that: The incident angle is θ i hour: in, Represents the viewing angle vector in the depth camera coordinate system, which is the vector from the depth camera position to the pixel i; Represents the normal vector at pixel i.

3. The method for evaluating the surface quality of initial shotcrete during underground gate excavation according to claim 1, characterized in that: The attenuation coefficient part is based on Lambertian scattering and specular reflection. The reflection attenuation model at different angles is constructed to describe V decay ,as follows: Where a is a constant, which is set according to the detection environment. In the environment of underground engineering with poor natural light, it is set to 2; n is the scattering coefficient, but when n = 1, it is regarded as a Lambertian scattering, and when n → + ∞, it is a specular reflection; that is, set:

4. The method for evaluating the surface quality of initial shotcrete during underground gate excavation according to claim 3 is characterized in that: The mechanism model of the depth camera is rewritten as follows: Among them, λ i (d i ,θ i ) is obtained by numerical analysis based on the collected data, as follows: Among them, A i , B i , C i , D i , E i , F i are the internal parameters of the depth camera, which together constitute the internal parameter matrix of the depth camera.

5. The method for evaluating the surface quality of initial shotcrete during underground gate excavation according to claim 1, characterized in that: The calculating based on the mechanism model of the depth camera to obtain the reflectivity at different pixel points in the image data of the surface to be detected includes: Preprocessing the image data of the surface to be detected by removing points outside the detection distance to obtain the preprocessed image data of the surface to be detected; The preprocessed image data of the surface to be detected is input into the mechanism model of the depth camera, and the reflectivity at different pixel points in the preprocessed image data of the surface to be detected is obtained.

6. The method for evaluating the surface quality of initial shotcrete during underground gate excavation according to claim 1, characterized in that: The initial spraying surface quality assessment process of the image data of the surface to be detected based on the reflectivity is as follows: Where S represents the standard deviation of the initial spray surface; R(θ i ) indicates that the incident angle at pixel i is θ i The reflectivity at ; M represents the mean; sum(i) represents the number of detected pixels; According to the requirements of the construction industry, a threshold value T0 is set. When the standard deviation of the initial sprayed surface is less than the set threshold value, S≤T0, the initial sprayed surface quality in the underground excavation operation meets the requirements, otherwise it does not meet the requirements.

7. A device for evaluating the surface quality of initial shotcrete in underground gate excavation operations based on a depth camera, characterized in that: The device is used to perform the method for evaluating the surface quality of initial shotcrete in underground gate excavation operations as described in any one of claims 1 to 6, and comprises: Modeling module: used for performing mechanism modeling processing on the depth camera based on the principle of the depth camera to obtain a mechanism model of the depth camera; Calibration module: used for performing camera calibration processing on the depth camera based on the mechanism model of the depth camera to obtain a calibrated depth camera; Acquisition module: used to acquire image data of the surface to be detected based on the calibrated depth camera to obtain image data of the surface to be detected; Calculation module: used for calculating and obtaining the reflectivity of different pixel points in the image data of the surface to be detected based on the mechanism model of the depth camera; Evaluation module: used for performing initial spraying surface quality evaluation processing on the image data of the surface to be detected based on the reflectivity to obtain an evaluation result.

Images