CN115753633A - Non-contact tunnel face surrounding rock water content detection method and system - Google Patents
Non-contact tunnel face surrounding rock water content detection method and system Download PDFInfo
- Publication number
- CN115753633A CN115753633A CN202211282400.7A CN202211282400A CN115753633A CN 115753633 A CN115753633 A CN 115753633A CN 202211282400 A CN202211282400 A CN 202211282400A CN 115753633 A CN115753633 A CN 115753633A
- Authority
- CN
- China
- Prior art keywords
- surrounding rock
- tunnel face
- tunnel
- face surrounding
- water content
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000011435 rock Substances 0.000 title claims abstract description 145
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 126
- 238000001514 detection method Methods 0.000 title claims abstract description 54
- 230000003595 spectral effect Effects 0.000 claims abstract description 42
- 238000001228 spectrum Methods 0.000 claims abstract description 25
- 239000011159 matrix material Substances 0.000 claims abstract description 24
- 238000000034 method Methods 0.000 claims description 29
- 230000006870 function Effects 0.000 claims description 16
- 238000005192 partition Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000012843 least square support vector machine Methods 0.000 claims description 7
- 238000013507 mapping Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 5
- 238000005457 optimization Methods 0.000 claims description 4
- 238000013135 deep learning Methods 0.000 claims description 3
- 241001365789 Oenanthe crocata Species 0.000 claims 1
- 230000004927 fusion Effects 0.000 claims 1
- 238000010276 construction Methods 0.000 abstract description 2
- 238000012545 processing Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 238000001035 drying Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000013508 migration Methods 0.000 description 3
- 230000005012 migration Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012706 support-vector machine Methods 0.000 description 3
- 230000005641 tunneling Effects 0.000 description 3
- 238000010521 absorption reaction Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000005553 drilling Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000009432 framing Methods 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 238000000691 measurement method Methods 0.000 description 2
- 238000004806 packaging method and process Methods 0.000 description 2
- 239000002352 surface water Substances 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000000701 chemical imaging Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000011900 installation process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000004080 punching Methods 0.000 description 1
- 230000003014 reinforcing effect Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
Abstract
The utility model belongs to the technical field of tunnel construction, concretely relates to non-contact tunnel face surrounding rock water content detection method and system, including: acquiring spectral information and image information of surrounding rocks on a tunnel face of a tunnel to be detected; identifying and judging the water outlet condition of the acquired image information to obtain the water outlet type of the surrounding rock of the tunnel face; selecting a face surrounding rock control point according to the obtained water outlet type of the face surrounding rock; fusing the acquired spectral information and image information of the tunnel face surrounding rock of the tunnel to be detected to obtain an image spectral matrix of the tunnel face surrounding rock; and completing the water content detection of the tunnel face surrounding rock according to the obtained image spectrum matrix of the face surrounding rock.
Description
Technical Field
The disclosure belongs to the technical field of tunnel construction, and particularly relates to a non-contact tunnel face surrounding rock water content detection method and system.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The water-rich property of the surrounding rock on the tunnel face is a key factor influencing the stability of the surrounding rock and is an important correction parameter in the classification of the surrounding rock. The tunnel surrounding rock generally has the characteristic of water absorption and softening, particularly the weakly consolidated surrounding rock is easy to generate structural deterioration and strength attenuation under the water absorption condition, so that the tunnel surrounding rock control method is a worldwide difficult problem for the tunnel surrounding rock control at present, if the dynamic water content of the tunnel surrounding rock can be accurately monitored, the tunnel can be driven and maintained according to the change conditions of the water content of different surrounding rocks, the tunnel surrounding rock damage condition is predicted in time, the tunnel surrounding rock loosening range is judged, tunnel surrounding rock supporting parameters and a reinforcing supporting scheme are adjusted in real time, early prevention and early management are performed on the tunnel surrounding rock roof fall accident, and safety during tunnel tunneling and service is improved.
According to the inventor, the existing tunnel surrounding rock water content detection method generally needs to be manually and periodically patrolled and checked and record the water outlet position and the water outlet condition, for the key water outlet area, punching operation needs to be further implemented, the water content detection equipment is installed to detect the water outlet quantity, the detection equipment needs to be checked and corrected before being used, instruments, sensors, materials and transmission wires are also used for carrying out continuity check in the installation process, the stability of the quality of the equipment is guaranteed, the installation and use processes of the water content detection equipment are very complicated and parts need to be regularly checked and replaced, and the overall detection working intensity is heavy. Under the existing detection conditions, the real-time performance of water content detection data is poor, the data processing efficiency is low, and the underground water permeation accident is difficult to forecast and early warn in time.
Disclosure of Invention
In order to solve the problems, the disclosure provides a non-contact tunnel face surrounding rock water content detection method and system, the water-rich degree of each area of tunnel surrounding rock is judged through image information, and for the area without obvious water-rich characteristics, the water content of the surrounding rock can be further detected through extracting spectral information.
According to some embodiments, a first aspect of the present disclosure provides a method for detecting moisture content of surrounding rock on a tunnel face of a non-contact tunnel, which adopts the following technical scheme:
a non-contact tunnel face surrounding rock water content detection method comprises the following steps:
acquiring spectral information and image information of surrounding rocks on a tunnel face of a tunnel to be detected;
identifying and judging the water outlet condition of the acquired image information to obtain the water outlet type of the surrounding rock of the tunnel face;
selecting a face surrounding rock control point according to the obtained water outlet type of the face surrounding rock;
fusing the acquired spectral information and image information of the tunnel face surrounding rock of the tunnel to be detected to obtain an image spectral matrix of the tunnel face surrounding rock;
and completing the water content detection of the tunnel face surrounding rock according to the obtained image spectrum matrix of the face surrounding rock.
As a further technical limitation, a preheated spectral imager is used for shooting images of surrounding rocks of the tunnel face to be detected in front of the tunnel, and spectral information and image information of the surrounding rocks of the tunnel face to be detected are obtained.
As a further technical limitation, before the obtained image information is subjected to the identification and judgment of the water outlet condition, the yolov5 target detection is adopted to detect the obtained image information of the surrounding rock of the tunnel face of the tunnel to be detected, and the detected image information is subjected to image partition to obtain a partition image.
Further, the acquired subarea images are identified and judged by adopting a TensorFlow deep learning algorithm, and the water outlet type of the tunnel face surrounding rock in the subarea images is judged through a preset classification and identification model;
the preset classification identification model adopts an increment-ResNet-V2 classification identification model, convolution kernels with different sizes are used for extracting image features, meanwhile, 1 x 1 convolution, 3 x 3 convolution and 5 x 5 convolution are calculated, output results are stacked along channel dimensions respectively, the stacked output results are transmitted to the next layer, and identification and judgment of the tunnel face surrounding rock water outlet type are completed.
As a further technical definition, the effluent types of the tunnel face surrounding rock include surface dry area, wet or drop effluent, strand or line effluent and spring effluent; and selecting tunnel face surrounding rock control points in the obtained surface drying image.
As a further technical limitation, in the process of fusing the acquired spectral information and the image information of the tunnel face surrounding rock of the tunnel to be detected, a mean normalization method is adopted to fuse the image information and the spectral information, and the characteristic values of the image information and the spectral information are respectively normalized to be within a specified interval, so that an image spectral matrix of the tunnel face surrounding rock is obtained.
As a further technical limitation, the specific process of detecting the water content of the tunnel face surrounding rock according to the obtained image spectrum matrix of the tunnel face surrounding rock comprises the following steps:
establishing a tunnel face surrounding rock water content sample library;
constructing a mapping relation between the water content sample library and the image spectrum matrix of the tunnel face surrounding rock;
constructing a least square support vector machine function according to the obtained mapping relation;
and (5) carrying out optimization solution on the obtained least square support vector machine function to complete the water content detection of the tunnel face surrounding rock.
According to some embodiments, a second aspect of the present disclosure provides a non-contact tunnel face surrounding rock water content detection system, which adopts the following technical solutions:
a non-contact tunnel face country rock water content detecting system includes:
the acquisition module is configured to acquire spectral information and image information of surrounding rocks on the tunnel face of the tunnel to be detected;
the recognition module is configured to recognize and judge the water outlet condition of the acquired image information to obtain the water outlet type of the face surrounding rock;
a selection module configured to select a face surrounding rock control point according to the obtained water outlet type of the face surrounding rock;
the calculation module is configured to fuse the acquired spectrum information and the acquired image information of the tunnel face surrounding rock of the tunnel to be detected to obtain an image spectrum matrix of the tunnel face surrounding rock;
and the detection module is configured to complete the water content detection of the tunnel face surrounding rock according to the obtained image spectrum matrix of the tunnel face surrounding rock.
According to some embodiments, a third aspect of the present disclosure provides a computer-readable storage medium, which adopts the following technical solutions:
a computer-readable storage medium, on which a program is stored, which, when executed by a processor, implements the steps in the non-contact tunnel face surrounding rock water content detection method according to the first aspect of the present disclosure.
According to some embodiments, a fourth aspect of the present disclosure provides an electronic device, which adopts the following technical solutions:
an electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor, where the processor executes the program to implement the steps in the method for detecting water content in surrounding rock of tunnel face of non-contact tunnel according to the first aspect of the present disclosure.
Compared with the prior art, the beneficial effect of this disclosure is:
the method provides a new method for judging different water-rich degrees of the face surrounding rock; the method comprises the steps of carrying out remote non-contact acquisition on tunnel surrounding rock images through a hyperspectral imager, judging and identifying surrounding rock water richness on a macroscopic scale through a pre-trained neural network according to image information, and further detecting water content in an area of a surface dry area through extracting spectral information at a control point. Compared with the traditional drilling measurement method, the method can be used for carrying out non-contact detection on the water content of the surrounding rock of the tunnel face in front of the tunnel, is simple to operate and easy to realize, has strong real-time water content detection data and high data processing efficiency, and can intelligently monitor the dynamic water content of the surrounding rock of the tunnel, thereby carrying out prediction and early warning on underground water permeation accidents in time and improving the safety during tunnel tunneling and service.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure and are not to limit the disclosure.
Fig. 1 is a flowchart of a method for detecting water content of surrounding rock on a tunnel face of a non-contact tunnel according to a first embodiment of the disclosure;
fig. 2 is a technical route diagram of a non-contact tunnel face surrounding rock water content detection method in a first embodiment of the disclosure;
FIG. 3 is a flowchart illustrating a migration training process of a target detection model according to a first embodiment of the disclosure;
FIG. 4 is a schematic view of water seepage and partition on a tunnel face according to one embodiment of the present disclosure;
FIG. 5 is a diagram of an inclusion-Resnet-V2 network architecture and a flowchart for implementing the same in one embodiment of the disclosure;
FIG. 6 is a schematic diagram of control point selection in a first embodiment of the disclosure;
fig. 7 is a block diagram of a non-contact tunnel face surrounding rock water content detection system in a second embodiment of the disclosure.
Detailed Description
The present disclosure is further described with reference to the following drawings and examples.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present disclosure. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict.
Example one
The embodiment of the disclosure introduces a non-contact tunnel face surrounding rock water content detection method.
As shown in fig. 1 and fig. 2, a non-contact tunnel face surrounding rock water content detection method includes:
acquiring spectral information and image information of surrounding rocks on a tunnel face of a tunnel to be detected;
identifying and judging the water outlet condition of the acquired image information to obtain the water outlet type of the surrounding rock of the tunnel face;
selecting a face surrounding rock control point according to the obtained water outlet type of the face surrounding rock;
fusing the acquired spectral information and image information of the tunnel face surrounding rock of the tunnel to be detected to obtain an image spectral matrix of the tunnel face surrounding rock;
and completing the water content detection of the tunnel face surrounding rock according to the obtained image spectrum matrix of the face surrounding rock.
Specifically, the present embodiment includes a data acquisition part and a data processing part; the data acquisition part acquires a tunnel face image containing spectral information through a hyperspectral imager; the data processing part mainly comprises 3 stages, wherein in the first stage, a target detection model is used for framing the position of a tunnel face image, and the framing result is combined to perform partition cutting on the target; the second stage is that the water outlet condition of the subareas is judged through a pre-trained palm surface water outlet type prediction model increment-ResNet-V2; and in the third stage, spectral information extraction is carried out on the subarea of which the water outlet condition is judged to be the surface drying area, and the water content is further measured by utilizing a partial least squares support vector machine model established in a laboratory in combination with the subarea spectral information and the image information.
As one or more embodiments, a spectral imager is used for shooting a tunnel front tunnel face to acquire a corresponding tunnel face image; before data acquisition, the hyperspectral imaging system needs to be preheated, so that the influence of the equipment moving process on the image acquisition quality is eliminated.
And performing target detection on the position of the palm surface by using a yolov5 target detection model which is subjected to migration training on the CIFAR-10 data set, and storing a detection result in a txt file in a coordinate mode. The migration training process is shown in fig. 3.
In one or more embodiments, the palm surface is cropped in 4 × 4 partitions according to the target detection file and the initial palm surface picture by using a self-programming procedure, and the partitions are schematically shown in fig. 4.
Establishing a palm surface water outlet sample database by utilizing the partition result, wherein the water outlet type comprises the following steps: (1) the surface dry area, (2) wet or drip-like effluent, (3) strand-like or linear running water, and (4) fountain-like effluent, the four effluent conditions are schematically shown in fig. 4, and the digital labels 1 to 4 represent the surface dry area, the wet or drip-like effluent, the strand-like or linear running water, the fountain-like effluent, and the like, respectively. And 4 labels such as digital labels 1-4 are used for marking and pre-classifying the subarea images, and the real water outlet condition of each subarea image only belongs to one category.
As one or more embodiments, an inclusion-ResNet-V2 network model is trained on a palm face water sample dataset using a TensorFlow deep learning framework. The database is divided into 8:2, the proportion is randomly distributed into two groups of a training set and a verification set, and the weights and the bias of the neural network can be gradually corrected in the training and verification process, so that the calculation precision is improved, and the occurrence of over-fitting and under-fitting conditions is reduced. The inclusion-ResNet-V2 network extracts image features using convolution kernels of different sizes, while computing 1 × 1, 3 × 3, 5 × 5 convolutions, and then stacking their output results along the channel dimension and passing on to the next layer, respectively. Different convolution operations and pooling operations of 1 × 1, 3 × 3, 5 × 5 and the like can better extract texture and threshold change information of the input image, and the rich water characteristics of the partitions can be better extracted by processing the operations in parallel and combining all the results. Storing and packaging the trained network model and related parameters; and inputting the partitioning result into the encapsulated inclusion-ResNet-V2 classification prediction model, and judging the water outlet condition of the partition. The inclusion-Resnet-V2 network architecture diagram and partition water-enrichment prediction process are shown in fig. 5.
As one or more embodiments, for the subarea whose identification result is the surface drying area, 16 control points (as shown in fig. 6) are set on the subarea, and the corresponding spectral curves at the control points are extracted, wherein each spectral curve has 200 wavelength points; and (3) storing R, G, B, H, S and I characteristic values in RGB and HSI color spaces at each control point, and then extracting 5 image texture information of energy, entropy, contrast, inverse difference moment and correlation of a gray level co-occurrence matrix, wherein the total number of the image texture information is 11. The spectral and image information are combined, i.e. 200 raw spectra and 11 image information for a total of 211 feature values. The characteristic values comprise image data and spectrum data, and the unit and the dimension of the characteristic values are different, namely the magnitude of the characteristic values are greatly different, so that the data weight is greatly different due to the large difference, and the prediction error is larger. In order to unify the magnitude of input data, the image and the spectral characteristic value are respectively normalized to [0,1 ] by adopting a mean value normalization method]The method specifically comprises the following steps:in the formula: x' is normalized image or spectral data; x is the initial image or spectral data; x is the number of max 、x min The minimum value and the maximum value of the image and the spectral characteristic value are respectively.
A sample library of different water contents of a typical rock body is established through indoor tests, and sample information comprises spectral characteristics, image characteristics and corresponding water content values. The spectral features are 200 wavelength point values of a typical rock spectral curve, the image features comprise R, G, B, H, S, I feature values in RGB and HSI color spaces, image texture information such as energy, entropy, contrast, inverse difference moment, correlation and the like of a gray level co-occurrence matrix and the like 11 image features, the spectral features and the image features are subjected to normalization processing and stored, and the normalization method adopts the mean value normalization processing method.
And establishing a partial least squares support vector machine model by using the water content value of the sample and the image and spectrum matrix of the sample to obtain a mapping relation between the water content and the image spectrum matrix.
Setting a data sample set { (x) 1 ,y 1 ),(x 2 ,y 2 ),…,(x t ,y t ) In which x is i ∈R n Normalized spectral and image feature values, y, for the sample i e.R, as the water content value of the sample, the least square support vector machine function for mapping the sample data set to the high-dimensional space is as follows:
in the formula (I), the compound is shown in the specification,for the non-linear mapping function, b is the deviation value and ω is the weight vector value. Through high dimensional spatial transformation, the optimization problem of the least squares support vector machine function can be transformed to solve the following equation:wherein the constraint condition of the conversion isIn the formula: gamma is a regularization parameter, and gamma is more than 0; e is a prediction error variable on the sample data set; e.g. of the type k Introducing an error variable for each sample; y is k To take into account the constraints of the bias and loss functions.
For the optimal solution, the Lagrangian transfer function L is in the form ofIn the formula, a k Is a lagrange multiplier.
The lagrangian function L (ω, b, e, a) is used to calculate the partial derivatives of the variables ω, b, e, a, and the following linear equation system can be obtained by using the KKT optimality condition:
by solving the linear equation set through calculation, the linear equation set can be obtained,wherein a = [ a ] 1 ,a 2 ,…,a N ];y=[y 1 ,y 2 ,…,y N ];I=[1,1,…,1];Is a kernel function.
The water content prediction function of the least square support vector machine can be finally expressed as
Different kernel functions are taken to form different least square support vector machine models, and radial basis functions are selectedAs the kernel function of the prediction model, the only parameters that need to be determined are the kernel parameter σ and the regularization parameter γ. The selection of two parameters is regarded as an optimization problem, and the preferred determination of sigma and gamma is carried out by adopting a genetic algorithm. Setting the calculation interval of sigma to [1, 100 ]]The calculation interval of γ is (0, 100)]The initial population scale of the genetic algorithm is set to be 20, the evaluation function adopts a mean square error reciprocal form, and the maximum iteration times are 50. And storing and packaging the water content prediction model after the parameters are optimized, and using the water content prediction model in the seventh step of water content detection identified as the surface drying subarea.
A new method is provided for judging different water-rich degrees of the face surrounding rock; the method comprises the steps of carrying out remote non-contact acquisition on tunnel surrounding rock images through a hyperspectral imager, judging and identifying surrounding rock water richness on a macroscopic scale through a pre-trained neural network according to image information, and further detecting water content in a region with a dry surface through extracting spectral information at a control point. Compared with the traditional drilling measurement method, the method can carry out non-contact detection on the water content of the surrounding rock of the tunnel face in front of the tunnel, is simple to operate and easy to realize, has strong real-time water content detection data and high data processing efficiency, and can intelligently monitor the dynamic water content of the surrounding rock of the tunnel, thereby forecasting and early warning the underground water-permeable accident in time and improving the safety during tunneling and service.
Example two
The second embodiment of the disclosure introduces a non-contact tunnel face surrounding rock water content detection system.
As shown in fig. 7, a non-contact tunnel face surrounding rock water content detection system includes:
the acquisition module is configured to acquire spectral information and image information of surrounding rocks on the tunnel face of the tunnel to be detected;
the recognition module is configured to recognize and judge the water outlet condition of the acquired image information to obtain the water outlet type of the face surrounding rock;
a selection module configured to select a face surrounding rock control point according to the obtained water outlet type of the face surrounding rock;
the calculation module is configured to fuse the acquired spectrum information and the acquired image information of the tunnel face surrounding rock of the tunnel to be detected to obtain an image spectrum matrix of the tunnel face surrounding rock;
and the detection module is configured to complete the water content detection of the tunnel face surrounding rock according to the obtained image spectrum matrix of the tunnel face surrounding rock.
The detailed steps are the same as those of the method for detecting the water content of the surrounding rock on the tunnel face of the non-contact tunnel provided in the first embodiment, and are not described herein again.
EXAMPLE III
The third embodiment of the disclosure provides a computer-readable storage medium.
A computer-readable storage medium, on which a program is stored, which when executed by a processor implements the steps in the non-contact tunnel face surrounding rock water content detection method according to the first embodiment of the present disclosure.
The detailed steps are the same as those of the method for detecting the water content of the surrounding rock on the tunnel face of the non-contact tunnel provided in the first embodiment, and are not described herein again.
Example four
The fourth embodiment of the disclosure provides an electronic device.
An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor, where the processor executes the program to implement the steps in the method for detecting moisture content of surrounding rock on tunnel face of a non-contact tunnel according to an embodiment of the present disclosure.
The detailed steps are the same as those of the method for detecting the water content of the surrounding rock on the tunnel face of the non-contact tunnel provided in the first embodiment, and are not described herein again.
The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure, and various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.
Although the present disclosure has been described with reference to specific embodiments, it should be understood that the scope of the present disclosure is not limited thereto, and those skilled in the art will appreciate that various modifications and changes can be made without departing from the spirit and scope of the present disclosure.
Claims (10)
1. A non-contact tunnel face surrounding rock water content detection method is characterized by comprising the following steps:
acquiring spectral information and image information of surrounding rocks on a tunnel face of a tunnel to be detected;
identifying and judging the water outlet condition of the acquired image information to obtain the water outlet type of the surrounding rock of the tunnel face;
selecting a face surrounding rock control point according to the obtained water outlet type of the face surrounding rock;
fusing the acquired spectral information and image information of the tunnel face surrounding rock of the tunnel to be detected to obtain an image spectral matrix of the tunnel face surrounding rock;
and completing the water content detection of the tunnel face surrounding rock according to the obtained image spectrum matrix of the face surrounding rock.
2. The non-contact tunnel face surrounding rock water content detection method as claimed in claim 1, characterized in that the preheated spectral imager is used for shooting the image of the surrounding rock of the tunnel face to be detected in front of the tunnel, and the spectral information and the image information of the surrounding rock of the tunnel face to be detected are obtained.
3. The non-contact tunnel face surrounding rock water content detection method as claimed in claim 1, wherein before the obtained image information is subjected to the identification and judgment of the water outlet condition, the yolov5 target detection is adopted to detect the obtained image information of the tunnel face surrounding rock to be detected, and the detected image information is subjected to image partition to obtain a partition image.
4. The non-contact tunnel face surrounding rock water content detection method as claimed in claim 3, characterized in that the obtained subarea image is identified and judged by adopting a TensorFlow deep learning algorithm, and the water outlet type of the face surrounding rock in the subarea image is judged by a preset classification identification model;
the preset classification identification model adopts an Incepti on-ResNet-V2 classification identification model, convolution kernels with different sizes are used for extracting image features, meanwhile, 1 x 1 convolution, 3 x 3 convolution and 5 x 5 convolution are calculated, output results are stacked along the channel dimension respectively, the stacked output results are transmitted to the next layer, and identification and judgment of the water outlet type of the face surrounding rock are completed.
5. The non-contact tunnel face surrounding rock water content detection method as claimed in claim 1, wherein the water outlet types of the tunnel face surrounding rock comprise surface dry area, wet or drop water, strand or line water and spring water; and selecting tunnel face surrounding rock control points in the obtained image of the surface dry area.
6. The non-contact tunnel face surrounding rock water content detection method as claimed in claim 1, characterized in that in the process of fusing the acquired spectrum information and image information of the tunnel face surrounding rock to be detected, the fusion of the image information and the spectrum information is performed by using a mean normalization method, and the characteristic values of the image information and the spectrum information are respectively normalized to be within a designated interval, so as to obtain the image spectrum matrix of the face surrounding rock.
7. The method for detecting the water content of the surrounding rock of the tunnel face in the non-contact manner as claimed in claim 1, wherein the specific process of detecting the water content of the surrounding rock of the tunnel face according to the obtained image spectrum matrix of the surrounding rock of the tunnel face comprises the following steps:
establishing a tunnel face surrounding rock water content sample library;
constructing a mapping relation between the water content sample library and the image spectrum matrix of the face surrounding rock;
constructing a least square support vector machine function according to the obtained mapping relation;
and (5) carrying out optimization solution on the obtained least square support vector machine function to complete the water content detection of the tunnel face surrounding rock.
8. The utility model provides a non-contact tunnel face country rock water content detecting system which characterized in that includes:
the acquisition module is configured to acquire spectral information and image information of surrounding rocks on the tunnel face of the tunnel to be detected;
the recognition module is configured to recognize and judge the water outlet condition of the acquired image information to obtain the water outlet type of the face surrounding rock;
a selection module configured to select a face surrounding rock control point according to the obtained water outlet type of the face surrounding rock;
the calculation module is configured to fuse the acquired spectrum information and the acquired image information of the tunnel face surrounding rock of the tunnel to be detected to obtain an image spectrum matrix of the tunnel face surrounding rock; (ii) a
And the detection module is configured to complete the water content detection of the tunnel face surrounding rock according to the obtained image spectrum matrix of the tunnel face surrounding rock.
9. A computer-readable storage medium on which a program is stored, the program, when being executed by a processor, implementing the steps in the non-contact tunnel face surrounding rock water content detection method according to any one of claims 1 to 7.
10. An electronic device comprising a memory, a processor and a program stored on the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method for detecting moisture content in surrounding rock of tunnel face of non-contact tunnel according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211282400.7A CN115753633B (en) | 2022-10-19 | 2022-10-19 | Non-contact tunnel face surrounding rock water content detection method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211282400.7A CN115753633B (en) | 2022-10-19 | 2022-10-19 | Non-contact tunnel face surrounding rock water content detection method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115753633A true CN115753633A (en) | 2023-03-07 |
CN115753633B CN115753633B (en) | 2024-06-11 |
Family
ID=85352148
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211282400.7A Active CN115753633B (en) | 2022-10-19 | 2022-10-19 | Non-contact tunnel face surrounding rock water content detection method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115753633B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102788752A (en) * | 2012-08-08 | 2012-11-21 | 江苏大学 | Non-destructive detection device and method of internal information of crops based on spectrum technology |
US20180085003A1 (en) * | 2016-07-27 | 2018-03-29 | Verifood, Ltd. | Spectrometry systems, methods, and applications |
AU2020102682A4 (en) * | 2020-10-12 | 2020-12-03 | Xianguang Wang | A prediction method of skarn deposit based on hyperspectral remote sensing images |
CN114067094A (en) * | 2021-11-11 | 2022-02-18 | 山东大学 | Multispectral camera system for rock drilling jumbo and intelligent surrounding rock identification method |
US20220112806A1 (en) * | 2020-10-13 | 2022-04-14 | Institute Of Rock And Soil Mechanics, Chinese Academy Of Sciences | Safety early warning method and device for full-section tunneling of tunnel featuring dynamic water and weak surrounding rock |
CN115019171A (en) * | 2022-06-09 | 2022-09-06 | 广西北投公路建设投资集团有限公司 | Non-contact surrounding rock fast partition grading method |
-
2022
- 2022-10-19 CN CN202211282400.7A patent/CN115753633B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102788752A (en) * | 2012-08-08 | 2012-11-21 | 江苏大学 | Non-destructive detection device and method of internal information of crops based on spectrum technology |
US20180085003A1 (en) * | 2016-07-27 | 2018-03-29 | Verifood, Ltd. | Spectrometry systems, methods, and applications |
AU2020102682A4 (en) * | 2020-10-12 | 2020-12-03 | Xianguang Wang | A prediction method of skarn deposit based on hyperspectral remote sensing images |
US20220112806A1 (en) * | 2020-10-13 | 2022-04-14 | Institute Of Rock And Soil Mechanics, Chinese Academy Of Sciences | Safety early warning method and device for full-section tunneling of tunnel featuring dynamic water and weak surrounding rock |
CN114067094A (en) * | 2021-11-11 | 2022-02-18 | 山东大学 | Multispectral camera system for rock drilling jumbo and intelligent surrounding rock identification method |
CN115019171A (en) * | 2022-06-09 | 2022-09-06 | 广西北投公路建设投资集团有限公司 | Non-contact surrounding rock fast partition grading method |
Non-Patent Citations (1)
Title |
---|
TOBIAS H. KURZ 等: "Hyperspectral image analysis of different carbonate lithologies (limestone, karst and hydrothermal dolomites): the Pozalagua Quarry case study (Cantabria, North-west Spain)", SEDIMENTOLOGY, vol. 59, no. 2, 20 September 2011 (2011-09-20) * |
Also Published As
Publication number | Publication date |
---|---|
CN115753633B (en) | 2024-06-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Nhat-Duc et al. | Automatic recognition of asphalt pavement cracks using metaheuristic optimized edge detection algorithms and convolution neural network | |
CN105303198B (en) | A kind of remote sensing image semisupervised classification method learnt from fixed step size | |
CN103258214B (en) | Based on the Classifying Method in Remote Sensing Image of image block Active Learning | |
CN111079602A (en) | Vehicle fine granularity identification method and device based on multi-scale regional feature constraint | |
CN107220603A (en) | Vehicle checking method and device based on deep learning | |
CN111178206A (en) | Building embedded part detection method and system based on improved YOLO | |
CN103345643B (en) | A kind of Classifying Method in Remote Sensing Image | |
CN106228125A (en) | Method for detecting lane lines based on integrated study cascade classifier | |
CN103400151A (en) | Optical remote-sensing image, GIS automatic registration and water body extraction integrated method | |
CN103426158B (en) | The method of two phase Remote Sensing Imagery Change Detection | |
CN107895136B (en) | Coal mine area identification method and system | |
CN114705689A (en) | Unmanned aerial vehicle-based method and system for detecting cracks of outer vertical face of building | |
CN111881958B (en) | License plate classification recognition method, device, equipment and storage medium | |
CN106096655A (en) | A kind of remote sensing image airplane detection method based on convolutional neural networks | |
CN111222545B (en) | Image classification method based on linear programming incremental learning | |
CN112613454A (en) | Electric power infrastructure construction site violation identification method and system | |
CN113836999A (en) | Tunnel construction risk intelligent identification method and system based on ground penetrating radar | |
CN104182767A (en) | Active learning and neighborhood information combined hyperspectral image classification method | |
CN111881970A (en) | Intelligent outer broken image identification method based on deep learning | |
CN113033315A (en) | Rare earth mining high-resolution image identification and positioning method | |
CN105205816A (en) | Method for extracting high-resolution SAR image building zone through multi-feature weighted fusion | |
CN115512247A (en) | Regional building damage grade assessment method based on image multi-parameter extraction | |
CN113378912B (en) | Forest illegal reclamation land block detection method based on deep learning target detection | |
CN117789038B (en) | Training method of data processing and recognition model based on machine learning | |
Katrojwar et al. | Design of Image based Analysis and Classification using Unmanned Aerial Vehicle |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |