CN117804368B - Tunnel surrounding rock deformation monitoring method and system based on hyperspectral imaging technology - Google Patents
Tunnel surrounding rock deformation monitoring method and system based on hyperspectral imaging technology Download PDFInfo
- Publication number
- CN117804368B CN117804368B CN202311816114.9A CN202311816114A CN117804368B CN 117804368 B CN117804368 B CN 117804368B CN 202311816114 A CN202311816114 A CN 202311816114A CN 117804368 B CN117804368 B CN 117804368B
- Authority
- CN
- China
- Prior art keywords
- surrounding rock
- mineral
- information
- map
- deformation
- 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.)
- Active
Links
- 239000011435 rock Substances 0.000 title claims abstract description 87
- 238000012544 monitoring process Methods 0.000 title claims abstract description 29
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000000701 chemical imaging Methods 0.000 title claims abstract description 20
- 238000005516 engineering process Methods 0.000 title claims abstract description 18
- 229910052500 inorganic mineral Inorganic materials 0.000 claims abstract description 92
- 239000011707 mineral Substances 0.000 claims abstract description 92
- 238000001228 spectrum Methods 0.000 claims abstract description 50
- 238000003384 imaging method Methods 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 230000003595 spectral effect Effects 0.000 claims description 23
- 238000012952 Resampling Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000003708 edge detection Methods 0.000 claims description 8
- 238000002329 infrared spectrum Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 230000004927 fusion Effects 0.000 claims description 4
- 239000003086 colorant Substances 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 abstract description 11
- 230000000694 effects Effects 0.000 abstract description 5
- 238000011160 research Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 230000001681 protective effect Effects 0.000 description 3
- 229910052736 halogen Inorganic materials 0.000 description 2
- 150000002367 halogens Chemical class 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000002310 reflectometry Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 239000003595 mist Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/16—Measuring arrangements characterised by the use of optical techniques for measuring the deformation in a solid, e.g. optical strain gauge
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/58—Extraction of image or video features relating to hyperspectral data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
- G06V10/806—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30132—Masonry; Concrete
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/06—Recognition of objects for industrial automation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Quality & Reliability (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a tunnel surrounding rock deformation monitoring method and system based on hyperspectral imaging technology, wherein the method comprises the following steps: acquiring imaging hyperspectral data of tunnel surrounding rock, and preprocessing the hyperspectral data to obtain spectrum information; inputting the spectrum information into a spectrum identification model for carrying out mineral information identification to generate a mineral map n; repeating the steps to obtain a mineral filling map n+1; describing the deformation degree of a target subarea in the mineral filling map n+1 relative to a reference subarea in the mineral filling map n by using a shape function, and obtaining surrounding rock deformation information; according to the invention, surrounding rock is monitored in real time through an imaging hyperspectral technology, spectrum matching is carried out on data of each pixel point, and map drawing is carried out on mineral distribution conditions. By analyzing the mineral map, the deformation effect of the rock is reflected by the shape, position and arrangement change of the minerals, so that the deformation condition of the surrounding rock is monitored.
Description
Technical Field
The invention relates to the technical field of surrounding rock deformation monitoring in tunnels, in particular to a method and a system for monitoring surrounding rock deformation of tunnels based on hyperspectral imaging technology.
Background
Surrounding rock deformation monitoring is helpful for timely finding and identifying deformation conditions of surrounding rock, and is important for success and safety of tunnel engineering. Studying the stress and strain distribution, deformation mechanisms, and geological processes of rock helps identify potential risks in order to take appropriate precautions. By timely monitoring and coping with surrounding rock deformation, high-quality completion of tunnel engineering can be ensured, and potential risks and cost are reduced.
The existing methods for monitoring the deformation of the surrounding rock comprise a total station, a geological radar, a strain gauge and the like, the methods are all used for acquiring the data of the deformation of the surrounding rock from a macroscopic angle, and certain limitations exist, for example, the measurement range of the total station is generally limited, and if a large-range area needs to be measured, the position of the instrument needs to be moved for multiple measurements; signal reflection of the geological radar is easily disturbed, so that interpretation and analysis of the geological radar are more difficult; the strain gauge is biased to measure strain in a fixed direction at a point, and global measurement cannot be achieved. A large number of researches show that the change of the stress of the surrounding rock of the tunnel can lead to the change of minerals in the surrounding rock, and a large amount of mechanisms and information capable of reflecting the deformation of the surrounding rock of the tunnel are reserved at the edge. For example, when a rock body is stressed in different directions, minerals can be stretched and sheared, and the shapes of the minerals are changed; upon fracture and crack formation, minerals may separate or tilt from their original location; under high pressure environment, the pressure effect can cause the change of molecular structure, and mineral phase change and the like are generated. These changes occur on a microscopic scale without causing significant changes in macroscopic shape, and existing methods for monitoring deformation of surrounding rock in a tunnel site have not mined information on this aspect.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides a tunnel surrounding rock deformation monitoring method and system based on a hyperspectral imaging technology, which can reflect the deformation effect of rock by means of the shape, position and arrangement changes of minerals, and realize monitoring of the deformation condition of surrounding rock.
The technical scheme of the invention is as follows:
In a first aspect of the invention, a method for monitoring deformation of surrounding rock of a tunnel based on hyperspectral imaging technology is provided, which comprises the following steps:
Acquiring imaging hyperspectral data of tunnel surrounding rock, and preprocessing the hyperspectral data to obtain spectrum information;
inputting the spectrum information into a spectrum identification model for carrying out mineral information identification to generate a mineral map n;
repeating the steps to obtain a mineral filling map n+1;
and describing the deformation degree of the target subarea in the mineral filling map n+1 relative to the reference subarea in the mineral filling map n by using a shape function, and obtaining surrounding rock deformation information.
In some embodiments of the invention, the imaging hyperspectral data of the tunnel surrounding rock comprises: surrounding rock near infrared-short wave infrared spectrum information and surrounding rock thermal infrared spectrum information which are respectively acquired by two sensors of a hyperspectral camera are subjected to resampling, and then the information acquired by the two sensors is subjected to resampling.
In some embodiments of the invention, the resampling comprises: the information obtained by the two sensors is geometrically corrected and registered to obtain a consistent data set.
In some embodiments of the present invention, preprocessing hyperspectral data to obtain spectral information includes:
Noise filtering processing is carried out on the data set obtained by resampling;
Extracting spectral features, and then carrying out feature level fusion on the spectral features;
and performing dimension reduction treatment on the fused spectrum characteristics to extract characteristic spectrums, and selecting important characteristic spectrums as spectrum information.
In some embodiments of the invention, the spectral recognition model is a trained and validated spectral recognition model, including sequence encoding, self-attention computation, integrated feature representation, and DNN classifier.
In some embodiments of the invention, the mineral map is obtained by: and marking different minerals identified by the spectrum identification model on the image by using different colors to generate a mineral map.
In some embodiments of the invention, the mineral map is enhanced with an edge detection algorithm that uses Sobel, prewitt or Canny edge detection algorithms.
In some embodiments of the present invention, describing deformation degree of a target subarea in the mineral map n+1 relative to a reference subarea in the mineral map n by using a shape function, to obtain surrounding rock deformation information, including:
extracting features from the mineral map n image, wherein the features are key points, edges or inflection points in the image, and taking the key points, edges or inflection points as points to be measured P 0;
in an image of a mineral map N, taking a point P 0 to be measured as a center point, selecting a region with a corresponding size of N x N as a reference sub-region, and carrying out numerical description on local features of the sub-region;
Searching a region most similar to the reference subregion in the image of the mineral filling map n+1 by utilizing the characteristic description of the reference subregion, and matching to find a most similar part, wherein the part is a target subregion and is the corresponding position of a to-be-measured point in the state n+1;
Based on the target subarea, quantifying the deformation of the to-be-measured point relative to the reference point through a shape function; analyzing and describing the shape difference of the target image relative to the reference image by using the calculation result of the shape function, thereby obtaining the description of surrounding rock deformation;
Repeating the above operation in the image of the mineral map n, finding the next point P 1 to be measured, performing correlation matching, and further calculating to obtain full-field strain information of the corresponding region of interest, thereby obtaining surrounding rock deformation information.
In some embodiments of the invention, the surrounding rock deformation information is analyzed, and if the surrounding rock deformation information exceeds a threshold value, an alarm is sent; if the threshold value is not exceeded, continuing monitoring.
In a second aspect of the present invention, there is provided a tunnel surrounding rock deformation monitoring system based on hyperspectral imaging technology, comprising:
A spectral information acquisition module configured to: acquiring imaging hyperspectral data of tunnel surrounding rock, and preprocessing the hyperspectral data to obtain spectrum information;
a mineral map generation module configured to: inputting the spectrum information into a spectrum identification model for carrying out mineral information identification to generate a mineral map n; repeating the steps to obtain a mineral filling map n+1;
The surrounding rock deformation information acquisition module is configured to: and describing the deformation degree of the target subarea in the mineral filling map n+1 relative to the reference subarea in the mineral filling map n by using a shape function, and obtaining surrounding rock deformation information.
One or more of the technical schemes of the invention has the following beneficial effects:
(1) According to the invention, surrounding rock is monitored in real time through an imaging hyperspectral technology, spectrum matching is carried out on data of each pixel point, and map drawing is carried out on mineral distribution conditions. By analyzing the mineral map, the deformation effect of the rock is reflected by the shape, position and arrangement change of the minerals, so that the deformation condition of the surrounding rock is monitored.
(2) The invention utilizes hyperspectral imaging technology to measure electromagnetic spectrum response of rock and mineral, and detects mineralogical characteristic change of microscopic scale. In the surrounding rock deformation process, tiny arrangement and position change which cannot be directly observed on a macroscopic scale can be captured and detected by adopting a hyperspectral imaging technology, so that the invention provides a novel monitoring means for surrounding rock deformation, and the tiny change of the surrounding rock deformation can be more comprehensively and accurately understood and monitored.
(3) The invention adopts non-contact test, has large data acquisition quantity, can monitor under the conditions of high-temperature environment, large engineering surface and the like, can trigger alarm when the deformation exceeds the threshold value, and timely informs corresponding management personnel or professionals to further check and evaluate.
(4) The invention can obtain not only space image information but also spectrum information, the advantage of the map unification can intuitively display the space distribution condition of minerals in different time domains, the contour line of the minerals is obvious, and the monitoring characteristic points can be directly obtained.
Drawings
FIG. 1 is a flow chart of a tunnel surrounding rock deformation monitoring method based on imaging hyperspectrum of the present invention;
Fig. 2 is a schematic diagram of deformation monitoring based on spectral mineral mapping in accordance with the present invention.
Detailed Description
The invention will be further described with reference to the drawings and examples.
Example 1
In an exemplary embodiment of the present invention, a method for monitoring deformation of surrounding rock of a tunnel based on hyperspectral imaging technology is provided, as shown in fig. 1, and includes the following steps:
Step one, acquiring imaging hyperspectral data of tunnel surrounding rock, and preprocessing the hyperspectral data to obtain spectrum information.
Specifically, the imaging hyperspectral data of the tunnel surrounding rock comprises: surrounding rock near infrared-short wave infrared spectrum information and surrounding rock thermal infrared spectrum information which are respectively acquired by two sensors of a hyperspectral camera are subjected to resampling, and then the information acquired by the two sensors is subjected to resampling.
In this embodiment, the manner of acquiring spectral data using the hyperspectral camera is as follows: the hyperspectral camera is moved to the position right in front of the research area, two sensors (one sensor acquires near infrared-short wave infrared spectrum information and the other sensor acquires thermal infrared spectrum information) are aligned with the research area at the same distance and angle, a halogen lamp is turned on, a light source is enabled to completely cover the research area, the white board is moved to the research area to conduct white board calibration, and the information can be used for subsequent data processing.
The hyperspectral camera is calibrated and set, so that clear and accurate spectrum data can be obtained by the equipment, the hyperspectral camera starts to scan the tunnel surrounding rock to obtain tunnel surrounding rock hyperspectral imaging data, the data are derived from two sensors, the data format of the data is a three-dimensional data body, the data comprise space image information and spectrum information of each pixel point, and the wave band range covers visible-near infrared and thermal infrared wave bands.
In a specific implementation manner of this embodiment, the core device for data acquisition is a hyperspectral camera, which has functions of setting camera parameters, such as exposure time, aperture, focal length, etc., and these parameters need to be adjusted according to a specific acquisition scene to obtain the best image quality; the hyperspectral camera is arranged on a tunnel trolley and matched with parts such as a special light source, a liftable bracket, a protective cover and a calibrated whiteboard; the special light sources are arranged at two sides of the hyperspectral camera, the power of one lamp is 300w, a plurality of lamps are respectively arranged at the left side and the right side, the distance between the hyperspectral camera and surrounding rock is adjusted according to the size of a researched area, the number of halogen lamps is further adjusted, and the light sources need to completely cover the researched area; the liftable support can control the height of the hyperspectral camera, so that hyperspectral image data of each position of the surrounding rock are obtained; the hyperspectral camera is arranged in the protective cover, the opposite part of the camera is quartz glass, the protective cover can prevent dust and water mist from damaging the hyperspectral camera, and the calibrated whiteboard is placed in a research area, so that the camera can completely capture the surface of the whiteboard; further, the device also comprises a control device which can control the horizontal movement of the tunnel trolley, the vertical movement of the liftable support, the special light source and the hyperspectral camera.
Resampling the information of the above sensors to ensure that the image information of the two sensors are spatially consistent so that they can be better combined together for subsequent data fusion, analysis, the resampling comprising: the information obtained by the two sensors is geometrically corrected and registered to obtain a consistent data set.
Further, the step of resampling specifically includes:
The hyperspectral imaging data of the two sensors are geometrically corrected and registered, wherein the geometric correction adopts whiteboard correction, and the whole data set is adjusted by using a known reference reflectivity value in a whiteboard image so as to eliminate errors caused by illumination change or uneven response of the sensors, correct spatial distortion in the image acquisition process and ensure that the corresponding relation between pixels and actual geographic positions is accurate.
The data of the two sensors after the correction are projected to the same coordinate system, then the spatial resolution of one image is adjusted to match the other image, so that the images have similar spatial resolution, and finally the images are cut to have the same size and area coverage, and a consistent data set is obtained.
In this embodiment, preprocessing hyperspectral data to obtain spectral information includes:
Noise filtering processing is carried out on the data set obtained by resampling; noise filtering can improve data quality and extract useful information, and common methods include smoothing filtering, principal component analysis, wavelet transformation and the like;
the hyperspectral data after pretreatment is a three-dimensional cube, spectral characteristics are extracted in a spectral dimension by taking each pixel point as a research object, and spectral characteristics of each pixel point are calculated, specifically, a target area is selected, an end member spectrum is found out based on the pixel points, and spectral characteristics of the end member spectrum of the target area are calculated;
Feature level fusion is carried out on the spectrum features, and features from different spectrum bands are combined to form a richer and comprehensive feature set;
and performing dimension reduction processing on the fused spectrum characteristics to extract characteristic spectrums, selecting important characteristic spectrums according to information quantity, correlation and the like, selecting the important characteristic spectrums as spectrum information, and improving the calculation efficiency of the model by reducing spectrum dimension.
And secondly, inputting the spectrum information into a spectrum identification model for mineral information identification, and generating a mineral filling figure n.
In this embodiment, the spectral recognition model is a trained and validated spectral recognition model, including sequence encoding, self-attention computation, integrated feature representation, and DNN classifier.
Specifically, the sequence encodes: mapping the reflectivity on the wave band into continuous vector representation by using an embedding layer to form embedded representation of the sequence, and finishing sequence coding;
Self-attention calculation: the self-attention mechanism is applied to calculate the attention weight of each wave band to other wave bands, so that the information of other positions in the sequence can be focused, and the capturing of the relation between different wave bands in the spectrum information is facilitated;
The integration feature represents: the attention weight obtained by self-attention calculation is weighted and summed with the original information to obtain an integrated characteristic representation, which can emphasize the information which has important contribution to mineral identification;
DNN classifier: inputting the integrated characteristics into a full-connection layer of DNN for final mineral identification;
The training process of the whole model comprises the steps of inputting spectral data into the model, obtaining a final mineral classification result through self-attention calculation and feature integration, and then carrying out error back propagation and parameter updating through comparison with a real label, wherein specific model parameters, embedding dimensions and hidden layer dimensions are required to be adjusted and optimized according to the mineral data of an actual tunnel address area.
The mineral map is obtained by: different minerals identified by the spectrum identification model are marked on the image by different colors to generate a mineral map, and the mineral map is marked as a state n, namely the mineral map n.
Collecting hyperspectral imaging data of surrounding rock again, and repeating the steps to obtain a mineral map n+1.
And enhancing the mineral contour line by adopting an edge detection algorithm for the mineral map, wherein the edge detection algorithm adopts Sobel, prewitt or Canny edge detection algorithm.
And thirdly, describing the deformation degree of the target subarea in the mineral filling map n+1 relative to the reference subarea in the mineral filling map n by using a shape function, and obtaining surrounding rock deformation information.
Specifically, as shown in fig. 2, the method comprises the following steps:
extracting a to-be-detected point: extracting features from the mineral map n image, wherein the features can be key points, edges or inflection points in the image, and taking the key points, edges or inflection points as points to be measured P 0;
Reference subregion search: in an image of a mineral map N, taking a point P 0 to be measured as a center point, selecting a region with a corresponding size of N x N as a reference sub-region, and carrying out numerical description on local features of the sub-region;
Feature matching, searching a target subarea: by utilizing the characteristic description of the reference subarea, the most similar area with the reference subarea is searched in the image of the mineral map n+1 for matching, and the matching is usually carried out by using a normalized cross-correlation method and the like. Finding out the most similar part, wherein the part is a target sub-area and is the corresponding position of the to-be-measured point in the state n+1;
And (3) calculating a shape function: based on the target subarea, quantifying the deformation of the to-be-measured point relative to the reference point through a shape function; these shape functions are typically based on changes in the movement, rotation, scaling, etc. of feature points, which can be calculated from the matched feature points (e.g. corner points, edge points or other features of the mineral demarcation).
Deformation analysis: analyzing and describing the shape difference of the target image relative to the reference image by using the calculation result of the shape function, thereby obtaining the description of surrounding rock deformation; repeating the above operation in the image of the mineral map n, finding the next point P 1 to be measured, performing correlation matching, and further calculating to obtain full-field strain information of the corresponding region of interest, thereby obtaining surrounding rock deformation information.
Further, analyzing the surrounding rock deformation information, and if the surrounding rock deformation information exceeds a threshold value, sending out an alarm to inform corresponding management personnel or professionals to carry out further examination and evaluation; if the threshold value is not exceeded, continuing monitoring.
According to the tunnel surrounding rock deformation monitoring method, surrounding rock is monitored in real time through an imaging hyperspectral technology, spectrum matching is conducted on data of each pixel point, and mapping is conducted on mineral distribution conditions. By analyzing the mineral map, the deformation effect of the rock is reflected by the shape, position and arrangement change of the minerals, so that the deformation condition of the surrounding rock is monitored.
Example 2
In an exemplary embodiment of the present invention, a tunnel surrounding rock deformation monitoring system based on hyperspectral imaging technology is provided, including:
A spectral information acquisition module configured to: acquiring imaging hyperspectral data of tunnel surrounding rock, and preprocessing the hyperspectral data to obtain spectrum information;
a mineral map generation module configured to: inputting the spectrum information into a spectrum identification model for carrying out mineral information identification to generate a mineral map n; repeating the steps to obtain a mineral filling map n+1;
The surrounding rock deformation information acquisition module is configured to: and describing the deformation degree of the target subarea in the mineral filling map n+1 relative to the reference subarea in the mineral filling map n by using a shape function, and obtaining surrounding rock deformation information.
It will be appreciated by those skilled in the art that the modules or steps of the invention described above may be implemented by general-purpose computer means, alternatively they may be implemented by program code executable by computing means, whereby they may be stored in storage means for execution by computing means, or they may be made into individual integrated circuit modules separately, or a plurality of modules or steps in them may be made into a single integrated circuit module. The present invention is not limited to any specific combination of hardware and software.
While the foregoing description of the embodiments of the present invention has been presented in conjunction with the drawings, it should be understood that it is not intended to limit the scope of the invention, but rather, it is intended to cover all modifications or variations within the scope of the invention as defined by the claims of the present invention.
Claims (6)
1. The tunnel surrounding rock deformation monitoring method based on the hyperspectral imaging technology is characterized by comprising the following steps of:
Acquiring imaging hyperspectral data of tunnel surrounding rock, and preprocessing the hyperspectral data to obtain spectrum information;
the imaging hyperspectral data of the tunnel surrounding rock comprises: surrounding rock near infrared-short wave infrared spectrum information and surrounding rock thermal infrared spectrum information which are respectively acquired by two sensors of a hyperspectral camera are resampled;
inputting the spectrum information into a spectrum identification model for carrying out mineral information identification to generate a mineral map n;
The mineral map is obtained by: marking different minerals identified by the spectrum identification model on the image by using different colors to generate a mineral map;
repeating the steps to obtain a mineral filling map n+1;
Carrying out enhanced mineral contour line on the mineral map by adopting an edge detection algorithm, wherein the edge detection algorithm adopts Sobel, prewitt or Canny edge detection algorithm;
describing the deformation degree of a target subarea in the mineral filling map n+1 relative to a reference subarea in the mineral filling map n by using a shape function, and obtaining surrounding rock deformation information; comprising the following steps:
extracting features from the mineral map n image, wherein the features are key points, edges or inflection points in the image, and taking the key points, edges or inflection points as points to be measured P 0;
in an image of a mineral map N, taking a point P 0 to be measured as a center point, selecting a region with a corresponding size of N x N as a reference sub-region, and carrying out numerical description on local features of the sub-region;
Searching a region most similar to the reference subregion in the image of the mineral filling map n+1 by utilizing the characteristic description of the reference subregion, and matching to find a most similar part, wherein the part is a target subregion and is the corresponding position of a to-be-measured point in the state n+1;
Based on the target subarea, quantifying the deformation of the to-be-measured point relative to the reference point through a shape function; analyzing and describing the shape difference of the target image relative to the reference image by using the calculation result of the shape function, thereby obtaining the description of surrounding rock deformation;
Repeating the above operation in the image of the mineral map n, finding the next point P 1 to be measured, performing correlation matching, and further calculating to obtain full-field strain information of the corresponding region of interest, thereby obtaining surrounding rock deformation information.
2. The hyperspectral imaging technique based tunnel surrounding rock deformation monitoring method as claimed in claim 1, wherein the resampling comprises: the information obtained by the two sensors is geometrically corrected and registered to obtain a consistent data set.
3. The method for monitoring deformation of tunnel surrounding rock based on hyperspectral imaging technology as claimed in claim 2, wherein preprocessing hyperspectral data to obtain spectral information comprises:
Noise filtering processing is carried out on the data set obtained by resampling;
Extracting spectral features, and then carrying out feature level fusion on the spectral features;
and performing dimension reduction treatment on the fused spectrum characteristics to extract characteristic spectrums, and selecting important characteristic spectrums as spectrum information.
4. The method for monitoring deformation of tunnel surrounding rock based on hyperspectral imaging technology as claimed in claim 1, wherein the spectral recognition model is a trained and validated spectral recognition model, including sequence coding, self-attention calculation, integrated feature representation and DNN classifier.
5. The method for monitoring deformation of tunnel surrounding rock based on hyperspectral imaging technology as claimed in claim 1, wherein the surrounding rock deformation information is analyzed, and if the deformation information exceeds a threshold value, an alarm is sent; if the threshold value is not exceeded, continuing monitoring.
6. A tunnel surrounding rock deformation monitoring system based on hyperspectral imaging technique, characterized in that a method according to any one of claims 1-5 is performed, comprising:
A spectral information acquisition module configured to: acquiring imaging hyperspectral data of tunnel surrounding rock, and preprocessing the hyperspectral data to obtain spectrum information;
a mineral map generation module configured to: inputting the spectrum information into a spectrum identification model for carrying out mineral information identification to generate a mineral map n; repeating the steps to obtain a mineral filling map n+1;
The surrounding rock deformation information acquisition module is configured to: and describing the deformation degree of the target subarea in the mineral filling map n+1 relative to the reference subarea in the mineral filling map n by using a shape function, and obtaining surrounding rock deformation information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311816114.9A CN117804368B (en) | 2023-12-26 | 2023-12-26 | Tunnel surrounding rock deformation monitoring method and system based on hyperspectral imaging technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311816114.9A CN117804368B (en) | 2023-12-26 | 2023-12-26 | Tunnel surrounding rock deformation monitoring method and system based on hyperspectral imaging technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117804368A CN117804368A (en) | 2024-04-02 |
| CN117804368B true CN117804368B (en) | 2024-09-20 |
Family
ID=90434289
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311816114.9A Active CN117804368B (en) | 2023-12-26 | 2023-12-26 | Tunnel surrounding rock deformation monitoring method and system based on hyperspectral imaging technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117804368B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118009914B (en) * | 2024-04-08 | 2024-06-11 | 上海中医药大学附属岳阳中西医结合医院 | Infrared spectrum-based intelligent moxibustion robot part temperature deformation monitoring method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108956482A (en) * | 2017-05-27 | 2018-12-07 | 核工业北京地质研究院 | A kind of high-spectrum remote-sensing method for quickly identifying at Effect of volcanic hydrothermal fluid activities center |
| CN114233397A (en) * | 2022-02-24 | 2022-03-25 | 中铁十二局集团山西建筑构件有限公司 | Tunnel construction rock burst early warning system based on air-land amphibious robot |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5499363B2 (en) * | 2010-07-23 | 2014-05-21 | 日本電信電話株式会社 | Image input device, image input method, and image input program |
| CN115758686A (en) * | 2022-11-03 | 2023-03-07 | 中铁第四勘察设计院集团有限公司 | Method for evaluating influence of near-field active fracture on tunnel surrounding rock |
| CN116481449A (en) * | 2023-04-26 | 2023-07-25 | 国家能源集团宁夏煤业有限责任公司 | Surrounding rock deformation determination method and device and surrounding rock deformation detection system |
| CN116935214B (en) * | 2023-06-27 | 2024-04-12 | 福建鼎旸信息科技股份有限公司 | Space-time spectrum fusion method for satellite multi-source remote sensing data |
-
2023
- 2023-12-26 CN CN202311816114.9A patent/CN117804368B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108956482A (en) * | 2017-05-27 | 2018-12-07 | 核工业北京地质研究院 | A kind of high-spectrum remote-sensing method for quickly identifying at Effect of volcanic hydrothermal fluid activities center |
| CN114233397A (en) * | 2022-02-24 | 2022-03-25 | 中铁十二局集团山西建筑构件有限公司 | Tunnel construction rock burst early warning system based on air-land amphibious robot |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117804368A (en) | 2024-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Guan et al. | Automated pixel-level pavement distress detection based on stereo vision and deep learning | |
| Liu et al. | A detection and recognition system of pointer meters in substations based on computer vision | |
| CN111855664B (en) | Adjustable three-dimensional tunnel defect detection system | |
| CN113516660B (en) | Visual positioning and defect detection method and device suitable for train | |
| CN110069972B (en) | Automatic detection of real world objects | |
| CN107615334B (en) | Object recognition device and object recognition system | |
| RU2609434C2 (en) | Detection of objects arrangement and location | |
| Fukuda et al. | Vision-based displacement sensor for monitoring dynamic response using robust object search algorithm | |
| Jahanshahi et al. | Unsupervised approach for autonomous pavement-defect detection and quantification using an inexpensive depth sensor | |
| EP3401671B1 (en) | Detection device and detection method | |
| CN111487320B (en) | Three-dimensional ultrasonic imaging method and system based on three-dimensional optical imaging sensor | |
| WO2006054425A1 (en) | Three-dimensional measuring instrument, three-dimensional measuring method, and three-dimensional measuring program | |
| Puente et al. | Automatic detection of road tunnel luminaires using a mobile LiDAR system | |
| CN117804368B (en) | Tunnel surrounding rock deformation monitoring method and system based on hyperspectral imaging technology | |
| CN112906750B (en) | Hyperspectral image-based material analysis method and system | |
| Feng et al. | Computer vision for structural dynamics and health monitoring | |
| CN112446852A (en) | Tunnel imaging plane display method and intelligent defect identification system | |
| CN110533649B (en) | Unmanned aerial vehicle general structure crack identification and detection device and method | |
| CN105023270A (en) | Proactive 3D stereoscopic panorama visual sensor for monitoring underground infrastructure structure | |
| Auer et al. | Characterization of facade regularities in high-resolution SAR images | |
| JP2019046295A (en) | Monitoring device | |
| WO2011054040A1 (en) | System and method for integration of spectral and 3-dimensional imaging data | |
| KR20210044127A (en) | Visual range measurement and alarm system based on video analysis and method thereof | |
| Park et al. | Creating a digital outcrop model by using hyper-spectrometry and terrestrial LiDAR | |
| CN114266835A (en) | Deformation monitoring control method and system for non-measuring camera |
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 |