CN116908214A - Tunnel construction defect detection method and system based on digital twinning - Google Patents
Tunnel construction defect detection method and system based on digital twinning Download PDFInfo
- Publication number
- CN116908214A CN116908214A CN202311181674.1A CN202311181674A CN116908214A CN 116908214 A CN116908214 A CN 116908214A CN 202311181674 A CN202311181674 A CN 202311181674A CN 116908214 A CN116908214 A CN 116908214A
- Authority
- CN
- China
- Prior art keywords
- tunnel
- region
- image
- information
- abnormal
- 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
- 238000010276 construction Methods 0.000 title claims abstract description 108
- 230000007547 defect Effects 0.000 title claims abstract description 90
- 238000001514 detection method Methods 0.000 title claims abstract description 24
- 230000002159 abnormal effect Effects 0.000 claims abstract description 80
- 238000012544 monitoring process Methods 0.000 claims abstract description 50
- 238000000034 method Methods 0.000 claims abstract description 29
- 238000005286 illumination Methods 0.000 claims description 53
- 230000000877 morphologic effect Effects 0.000 claims description 42
- 230000005856 abnormality Effects 0.000 claims description 27
- 230000003213 activating effect Effects 0.000 claims description 25
- 238000013507 mapping Methods 0.000 claims description 21
- 238000004364 calculation method Methods 0.000 claims description 20
- 230000009471 action Effects 0.000 claims description 16
- 238000013461 design Methods 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 11
- 238000011156 evaluation Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 5
- 230000003068 static effect Effects 0.000 claims description 4
- 238000005192 partition Methods 0.000 claims description 2
- 238000012549 training Methods 0.000 abstract description 7
- 238000012545 processing Methods 0.000 abstract description 4
- 238000012876 topography Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000000638 solvent extraction Methods 0.000 description 3
- 230000001360 synchronised effect Effects 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000001960 triggered effect Effects 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/9515—Objects of complex shape, e.g. examined with use of a surface follower device
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- 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
- G06T7/0008—Industrial image inspection checking presence/absence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/9515—Objects of complex shape, e.g. examined with use of a surface follower device
- G01N2021/9518—Objects of complex shape, e.g. examined with use of a surface follower device using a surface follower, e.g. robot
Landscapes
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Immunology (AREA)
- General Health & Medical Sciences (AREA)
- Biochemistry (AREA)
- Analytical Chemistry (AREA)
- Pathology (AREA)
- Chemical & Material Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a tunnel construction defect detection method and system based on digital twinning, which relate to the technical field of data processing, and the method comprises the following steps: the method comprises the steps of receiving positioning information of a first tunnel region, downloading a first tunnel region reference image, performing image capturing to generate a first tunnel region monitoring image, receiving the first tunnel region reference image and the first tunnel region monitoring image, generating an image abnormal feature region and a region abnormal coefficient, obtaining a safety risk level, generating dangerous identification information for the image abnormal feature region when the safety risk level is greater than or equal to a safety risk level threshold value, synchronizing the dangerous identification information to a tunnel digital twin model, and sending the dangerous identification information to a user side, so that the defect monitoring of tunnel construction defect in the prior art is solved, the technical problem that accuracy is insufficient due to the fact that training data and actual scene data are deviated is solved, the reference image with high scene combination degree is obtained by utilizing the digital twin model, and defect monitoring with high accuracy is achieved.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a tunnel construction defect detection method and system based on digital twinning.
Background
Tunnel construction is a complex construction project in which many factors, such as geological conditions, hydrologic conditions, and possible earthquakes, etc., need to be considered. The most important in tunnel construction is to ensure the safety and stability of tunnel construction. In order to monitor the safety condition of tunnel construction, tunnel construction monitoring technology has been subjected to long-term research and development. The traditional tunnel monitoring technology is based on experience, and mostly judges geological conditions and tunnel stability according to experience and field observation in engineering construction. The main monitoring content of the conventional technology comprises tunnel geological conditions, tunnel inclination and displacement, subsidence of the ground around the tunnel and the like. The traditional technology has a plurality of defects such as long monitoring period, low data precision, and the judgment of the monitoring result needs to consider the influence of factors such as construction process, monitoring visual angle and the like,
the method solves the technical problems of insufficient accuracy caused by the fact that the abnormality identification is carried out on tunnel construction defect monitoring in the prior art, but the deviation between training data and actual scene data is caused, and achieves the defect monitoring with higher accuracy by obtaining a reference image with higher scene combination degree by utilizing a digital twin model.
Disclosure of Invention
The application provides a tunnel construction defect detection method and system based on digital twinning, which are used for solving the technical problems that in the prior art, abnormal recognition is carried out on tunnel construction defect monitoring, but the accuracy is insufficient due to deviation between training data and actual scene data.
In view of the above problems, the application provides a tunnel construction defect detection method and system based on digital twinning.
In a first aspect, the present application provides a method for detecting a tunnel construction defect based on digital twinning, the method comprising: when a tunnel construction defect identification robot runs to a first tunnel section, receiving first tunnel region positioning information, wherein a first tunnel region belongs to the first tunnel section; according to the first tunnel region positioning information, downloading a first tunnel region reference image from a tunnel digital twin model of a software processing console; acquiring shooting action characteristics of the first tunnel region reference image, and controlling a vision module of the tunnel construction defect recognition robot to perform image capturing so as to generate a first tunnel region monitoring image; activating a first twin node embedded in an edge processor of the tunnel construction defect identification robot to receive the first tunnel region reference image, and activating a second twin node of the edge processor to receive the first tunnel region monitoring image to generate an image abnormal characteristic region and a region abnormal coefficient; when the regional abnormal coefficient is larger than or equal to an abnormal coefficient critical value, the image abnormal characteristic region and the regional abnormal coefficient are sent to the software processing console for security risk association, and a security risk level is obtained; and when the security risk level is greater than or equal to a security risk level threshold, generating dangerous identification information for the image abnormal characteristic region, synchronizing the dangerous identification information to the tunnel digital twin model, and sending the dangerous identification information to a user side.
In a second aspect, the present application provides a digital twinning-based tunnel construction defect detection system, the system comprising: the information receiving module is used for receiving positioning information of a first tunnel area when the tunnel construction defect identification robot runs to the first tunnel section, wherein the first tunnel area belongs to the first tunnel section; the reference image module is used for downloading a first tunnel region reference image from a tunnel digital twin model of the software processing console according to the first tunnel region positioning information; the image capturing module is used for acquiring shooting action characteristics of the first tunnel region reference image, controlling a vision module of the tunnel construction defect recognition robot to capture images and generating a first tunnel region monitoring image; the abnormality module is used for activating a first twin node embedded in an edge processor of the tunnel construction defect identification robot to receive the first tunnel region reference image, and activating a second twin node of the edge processor to receive the first tunnel region monitoring image so as to generate an image abnormality characteristic region and a region abnormality coefficient; the security risk association module is used for sending the image abnormal characteristic region and the region abnormal coefficient to the software processing platform for security risk association when the region abnormal coefficient is greater than or equal to an abnormal coefficient critical value, and obtaining a security risk level; the first judging module is used for generating dangerous identification information for the image abnormal characteristic area when the security risk level is greater than or equal to a security risk level threshold value, synchronizing the dangerous identification information to the tunnel digital twin model and sending the dangerous identification information to a user side.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
the application provides a tunnel construction defect detection method and system based on digital twinning, relates to the technical field of data processing, solves the technical problems that in the prior art, abnormality identification is carried out on tunnel construction defect monitoring, but the accuracy is insufficient due to deviation between training data and actual scene data, and achieves the defect monitoring with higher accuracy by obtaining a reference image with higher scene combination degree by using a digital twinning model.
Drawings
FIG. 1 is a schematic flow diagram of a tunnel construction defect detection method based on digital twinning;
FIG. 2 is a schematic flow chart of a method for detecting a defect in a tunnel construction based on digital twinning to obtain positioning information of a first tunnel region;
FIG. 3 is a schematic flow chart of a process for generating a first tunnel region reference image in a tunnel construction defect detection method based on digital twinning;
FIG. 4 is a schematic flow chart of the regional anomaly coefficients in the tunnel construction defect detection method based on digital twinning;
fig. 5 is a schematic structural diagram of a tunnel construction defect detection system based on digital twinning.
Reference numerals illustrate: the system comprises an information receiving module 1, a reference image module 2, an image capturing module 3, an abnormality module 4, a security risk association module 5 and a first judging module 6.
Detailed Description
The application provides a tunnel construction defect detection method and system based on digital twinning, which are used for solving the technical problems that in the prior art, the defect detection of the tunnel construction is abnormal and identified, but the training data and the actual scene data have deviation, so that the accuracy is insufficient.
Embodiment one:
as shown in fig. 1, an embodiment of the present application provides a method for detecting a tunnel construction defect based on digital twinning, which is applied to a system for detecting a tunnel construction defect based on digital twinning, the system includes a software processing console and a hardware execution terminal, the hardware execution terminal includes a tunnel construction defect identification robot and a user terminal, and the method includes:
step S100: when a tunnel construction defect identification robot runs to a first tunnel section, receiving first tunnel region positioning information, wherein a first tunnel region belongs to the first tunnel section;
further, as shown in fig. 2, step S100 of the present application further includes:
Step S110: uploading a construction design original image and a quality design standard of a preset tunnel by a user side, and constructing the tunnel digital twin model of the preset tunnel;
step S120: positioning the preset tunnel in a grid space coordinate system, and synchronizing to the tunnel digital twin model;
step S130: and according to the Beidou positioning module of the tunnel construction defect identification robot, acquiring positioning information of the identification robot, synchronizing the positioning information to the tunnel digital twin model, and converting the robot into a static state when the tunnel construction defect identification robot runs to the first tunnel section to acquire the positioning information of the first tunnel region.
Specifically, the digital twinning-based tunnel construction defect detection method is applied to a digital twinning-based tunnel construction defect detection system, the digital twinning-based tunnel construction defect detection system comprises a software processing console and a hardware execution terminal, the hardware execution terminal comprises a tunnel construction defect identification robot and a user side, and the software processing console and the hardware execution terminal are used for detecting and collecting tunnel construction defect parameters.
When the tunnel construction defect recognition robot runs to a first tunnel section, the first tunnel section is used for receiving the positioning information of the first tunnel area, wherein the first tunnel section is used for selecting one section from a plurality of tunnel sections with equal length as the first tunnel section after the whole tunnel is uniformly divided into the plurality of tunnel sections with equal length, and the first tunnel area belongs to the first tunnel section,
uploading construction design original pictures and quality design standards of a preset tunnel through a user side, wherein the construction design original pictures and the quality design standards of the preset tunnel refer to theoretical values of the construction design original pictures and the quality design standards of the tunnel before construction, on the basis, a tunnel digital twin model of the preset tunnel, namely a theoretical tunnel digital twin model, is built in an equal ratio with the construction design original pictures and the quality design standards of the tunnel, further, the preset tunnel is positioned in a grid space coordinate system, the origin of coordinates of the grid space coordinate system is positioned at the center of the preset tunnel, the Z axis points to one end of the preset tunnel facing the sky, the X axis points to the port of any end of the preset tunnel and is perpendicular to the Z axis, the direction of the Y axis is one end of the preset tunnel perpendicular to the X axis, the Y axis is perpendicular to the Z axis, and finally, all coordinate points of the preset tunnel are positioned in the built network space coordinate system, and synchronizing the positioned coordinate points to a tunnel digital twin model, further, according to a Beidou positioning module in the tunnel construction defect recognition robot, the Beidou positioning module refers to a navigation positioning system in a first tunnel area, acquires positioning information of the recognition robot in the first tunnel area and synchronizes the positioning information into the tunnel digital twin model, and when the positioning information of the defect recognition robot is synchronized to the tunnel digital twin model, the defect recognition robot in the tunnel digital twin model can be visualized as a small red point for displaying, finally, when the tunnel construction defect recognition robot runs to a first tunnel section, the motion state of the robot is converted into a static state, and the positioning information of the first tunnel area is synchronously acquired according to the positioning information of the robot, and defect detection is carried out for later-stage tunnel construction based on digital twin as an important reference basis.
Step S200: according to the first tunnel region positioning information, downloading a first tunnel region reference image from a tunnel digital twin model of a software processing console;
further, as shown in fig. 3, step S200 of the present application further includes:
step S210: activating a photosensitive module of the tunnel construction defect identification robot to acquire sensing illumination vector information, wherein the illumination vector information comprises light intensity sensing information and illumination direction sensing information;
step S220: synchronizing the light intensity sensing information and the light direction sensing information to the tunnel digital twin model based on the first tunnel region positioning information, generating the first tunnel region reference image, including,
when the first tunnel region positioning information is received, a first tunnel region initial image is acquired, wherein the first tunnel region initial image has initial illumination vector information;
step S230: activating an image encoder embedded in the tunnel digital twin model, encoding the initial image of the first tunnel region based on the initial illumination vector information and the sensing illumination vector information, and generating the reference image of the first tunnel region.
Further, step S230 of the present application includes:
step S231: acquiring a first tunnel region identification image, wherein the first tunnel region identification image comprises a sensing illumination record vector;
step S232: mapping the initial illumination vector information, the sensing illumination record vector and the first tunnel region initial image through a convolution topological network to generate a first tunnel region coded image;
step S233: performing similarity evaluation on the first tunnel region coding image and the first tunnel region identification image to obtain a mapping loss amount, wherein the mapping loss amount is a similarity negative value;
step S234: and when the mapping loss amount is continuously preset for times smaller than the mapping preset loss amount, generating the image encoder.
Specifically, in order to ensure accuracy of detecting a tunnel construction defect in a later stage, downloading a reference image of a first tunnel region from a digital twin model of a tunnel in a software processing center included in a system according to first tunnel region positioning information is required, namely activating a photosensitive module included in a tunnel construction defect identification robot, wherein the photosensitive module is used for recording a phenomenon that chemical or physical properties change under irradiation of light in the tunnel, so as to acquire sensing illumination vector information through the photosensitive module, wherein the illumination vector information can include light intensity sensing information in the tunnel and sensing information of an illumination direction, the light intensity sensing information is sensing information for marking illumination intensity in the tunnel, the sensing information of the illumination direction is sensing information for marking the direction of the illumination in the tunnel, and the light intensity sensing information and the illumination direction sensing information can be represented by an angle.
Further, the image encoder embedded in the digital twin model of the tunnel is activated, firstly, the image encoder acquires an identification image in a first tunnel region, the identification image of the first tunnel region comprises a sensing illumination record vector, the sensing illumination record vector refers to light intensity information and illumination direction information which are sensed and recorded by a photosensitive module, further, initial illumination vector information, the sensing illumination record vector and a first tunnel region initial image are mapped through a convolution topology network, the initial illumination vector information acquired by the sensing module is based on the initial illumination vector information, the initial illumination vector information is equally divided based on the sensing illumination record vector on the basis of the acquired initial illumination vector information, meanwhile, a first region in the initial illumination vector information equally divided is set as a starting point, namely the acquired first region is identified as a zero region, traversal is started from the first region, the acquired information in each region is matched with the sensing record vector, thus the initial tunnel region illumination information is generated, and simultaneously, the initial tunnel region illumination vector information is mapped with the first initial region image, namely the first initial illumination image is mapped in the first initial region, the first initial image is coded according to a first initial image value, the first initial image is further coded, the tunnel value is taken in the first initial image, the first initial image is mapped in the first initial image region, the first initial image is coded, the first initial image is evaluated in the first initial image region, and the first initial image is mapped in the first initial image region, and the first initial image is similar to the initial image, the similarity evaluation is used for measuring the similarity between the first tunnel region coded image and the first tunnel region identification image, the higher the similarity is, the higher the similarity evaluation value is, when the similarity evaluation value is negative, the negative value is used as the mapping loss amount when mapping is carried out, and when the continuous preset times of the mapping loss amount is smaller than the mapping preset loss amount, the image encoder is generated, wherein the mapping preset loss amount is preset by a related technician according to a loss function, the initial image of the first tunnel region is coded by the image encoder based on initial illumination vector information and sensing illumination vector information, the initial image of which the image coding is completed is used as the first tunnel region reference image, and further defect detection is guaranteed for realizing tunnel construction based on digital twin.
Step S300: acquiring shooting action characteristics of the first tunnel region reference image, and controlling a vision module of the tunnel construction defect recognition robot to perform image capturing so as to generate a first tunnel region monitoring image;
specifically, in order to better detect the defects in the tunnel construction, it is first necessary to extract the action characteristics of the first tunnel region reference image downloaded from the digital twin model of the tunnel in the software processing center, where the shooting action characteristics refer to the shooting angle, the shooting height, and the like when the first tunnel region reference image is shot, so as to control the vision module in the tunnel construction defect recognition robot to shoot the first tunnel region reference image, automatically acquire the picture resources on the first tunnel region through the web crawler technology, and monitor the first tunnel region according to the acquired picture resources, thereby generating the first tunnel region monitoring image and tamping the defect detection for the subsequent implementation of the digital twin-based tunnel construction.
Step S400: activating a first twin node embedded in an edge processor of the tunnel construction defect identification robot to receive the first tunnel region reference image, and activating a second twin node of the edge processor to receive the first tunnel region monitoring image to generate an image abnormal characteristic region and a region abnormal coefficient;
Further, as shown in fig. 4, step S400 of the present application further includes:
step S410: grouping the first tunnel region monitoring images and the first tunnel region reference images according to the shooting action characteristics, and obtaining a plurality of groups of first tunnel region monitoring images and first tunnel region reference images with consistent shooting action characteristics;
step S420: performing morphological feature extraction on the first tunnel region reference image by using the first twin node to obtain first morphological feature information;
step S430: performing morphological feature extraction on the first tunnel region monitoring image by using the second twin node to obtain second morphological feature information;
step S440: performing deviation calculation on the first morphological feature information and the second morphological feature information to obtain the image abnormal feature region with the feature deviation distance being greater than or equal to a deviation distance threshold;
step S450: and setting the average value of the characteristic deviation distances of the image abnormal characteristic areas as the area abnormal coefficient.
Further, step S440 of the present application includes:
step S441: acquiring the first morphological feature information and the second morphological feature information of a first pixel point, and performing Euclidean distance calculation to acquire a first feature deviation distance;
Step S442: acquiring the first morphological feature information and the second morphological feature information of the N pixel point, and performing Euclidean distance calculation to acquire an N feature deviation distance;
step S443: traversing the first characteristic deviation distance to the Nth characteristic deviation distance, and adding pixel points which are larger than or equal to the deviation distance threshold value into an abnormal pixel point set;
step S444: and traversing the pixel point position information of the abnormal pixel point set to partition according to the region continuous distance threshold value, and generating the image abnormal characteristic region.
Specifically, the downloaded first tunnel region reference image and the grabbed first tunnel region monitoring image are used as basic image data for detecting the tunnel construction defects, and firstly an edge processor embedded in the tunnel construction defect identification robot is activated, wherein the edge processor is independent of a calculation module outside a control center of the robot.
Further, grouping the first tunnel region monitoring image and the first tunnel region reference image according to different shooting heights and shooting angles in shooting action features, namely dividing the first tunnel region monitoring image and the first tunnel region reference image with the same shooting height and shooting angle into the same group, acquiring a plurality of groups of first tunnel region monitoring images and first tunnel region reference images with consistent shooting action features, simultaneously carrying out profile feature extraction on the first tunnel region reference images by using a first twin node, acquiring first profile feature information of the first tunnel region reference images, carrying out profile feature extraction on the first tunnel region monitoring images by using a second twin node, acquiring second profile feature information of the first tunnel region monitoring images, wherein the first twin node and the second twin node are twin network models, training a plurality of images from the first tunnel region reference images by using the first twin node, at least comprising two categories in the selected plurality of images, calculating a square function of second norms of the selected plurality of images, judging a profile feature difference by using the first twin node, carrying out profile feature extraction on the first profile feature information of the first tunnel region reference images, and finally carrying out profile feature extraction on the first profile feature information of the first tunnel region reference images by using the second twin node, and finally carrying out profile feature extraction on the first profile feature information.
And training a plurality of images selected from the first tunnel region monitoring image through the second twin node, wherein the selected plurality of images at least comprise two categories, so that the square of the two norms of the vector difference of the selected plurality of images is calculated, the sample distance of the plurality of images is judged through a loss function, the first tunnel region monitoring image is finally input into the trained second twin node to extract the morphological features of the first tunnel region monitoring image, and the extracted morphological features of the tunnel are recorded as second morphological feature information.
Further, performing deviation calculation on the first morphology feature information and the second morphology feature information, namely taking the sum of absolute values of deviations of single measured values and average values, wherein the relative standard deviation is the percentage of the standard deviation to the average value, so that feature deviation distances of the first morphology feature information and the second morphology feature information are determined according to a deviation calculation result, the feature deviation distances are compared with a threshold value according to the deviation distances, the threshold value of the deviation distances is defined according to the upper limit value of historical deviation distance data and the lower limit value of the historical deviation distance data, further image feature areas with the feature deviation distances larger than or equal to the deviation distance threshold value are marked as image abnormal feature areas, and the first morphology feature information and the second morphology feature information of the first pixel point are subjected to Euclidean distance calculation according to the following formula:
；
Wherein d is the Euclidean distance between the first morphological feature information and the second morphological feature information of the first pixel point,for the data of the first topography characteristic x point of the first pixel point in the network space coordinate system,/for the data of the first topography characteristic x point of the first pixel point>For the data of the first topography feature signal y point of the first pixel point in the network space coordinate system,/for the data of the first topography feature signal y point of the first pixel point in the network space coordinate system>For the data of the first topography feature signal z point of the first pixel point in the network space coordinate system,/for the data of the first topography feature signal z point of the first pixel point in the network space coordinate system>For the data of the second morphological feature x-point of the first pixel point in the network space coordinate system +.>For the data of the second morphological feature signal y point of the first pixel point in the network space coordinate system +.>And determining the first characteristic deviation distance of the first pixel point according to the calculated Euclidean distance for the data of the second morphological characteristic signal z point of the first pixel point in the network space coordinate system.
Performing iterative computation on the basis until the iteration is performed on the N pixel points, simultaneously acquiring first morphological feature information and second morphological feature information of the N pixel points, performing Euclidean distance computation, acquiring N feature deviation distances, wherein N can be the number of all the pixel points in the image, further, sequentially traversing the first feature deviation distances until the N feature deviation distances, comparing the feature deviation distances accessed by each traversing with a deviation distance threshold, adding the pixel points corresponding to the feature deviation distances which are greater than or equal to the deviation distance threshold into an abnormal pixel point set, and finally traversing and partitioning the pixel point position information of the abnormal pixel point set according to the region continuous distance threshold, namely, regarding the pixel point aggregation of the abnormal pixel point set as the same region, acquiring a plurality of continuous regions, determining the image abnormal feature regions according to the continuous regions, calculating the mean value of the feature deviation distances of the image abnormal feature regions, and recording the mean value of the calculated mean value as a tunnel defect detection factor based on the fact that the number is defined by the tunnel is realized.
Step S500: when the regional abnormal coefficient is larger than or equal to an abnormal coefficient critical value, the image abnormal characteristic region and the regional abnormal coefficient are sent to the software processing console for security risk association, and a security risk level is obtained;
further, the step S500 of the present application further includes:
step S510: searching in the tunnel engineering big data according to the tunnel digital twin model, the image abnormal characteristic region and the region abnormal coefficient to generate M pieces of construction transaction data, wherein the M pieces of construction transaction data comprise accident triggering frequency, M is more than or equal to 500, and M is an integer;
step S520: calculating the safety risk coefficient, wherein the safety risk coefficient is the ratio of the accident triggering frequency to the M pieces of construction business data;
step S530: and activating a security risk level table to process the security risk coefficient to acquire the security risk level, wherein the security risk coefficient and the security risk level in the security risk level table are in one-to-one correspondence.
Specifically, judging the regional abnormality coefficient obtained by calculation, when the regional abnormality coefficient is greater than or equal to an abnormality coefficient critical value, sending an image abnormality characteristic region and the regional abnormality coefficient to a software processing center for security risk association, wherein the abnormality coefficient critical value refers to M business data based on the regional abnormality coefficient in data processing, analysis or calculation, when the regional abnormality coefficient reaches the upper limit or the lower limit of an abnormality threshold value, the threshold value is called an abnormality data critical value, further, searching in tunnel engineering big data according to a tunnel digital twin model, the image abnormality characteristic region and the regional abnormality coefficient, searching in the tunnel engineering big data, and obtaining a security risk table according to a new security risk table by using a comparison method and a technical means of tunnel abnormality, finding out information related to the tunnel digital twin model, the image abnormality characteristic region and the regional abnormality coefficient from the tunnel engineering big data according to a tunnel abnormality cue and a rule, and recording the found related information as M business data, wherein the M business data comprises an accident triggering frequency, namely the frequency of occurrence of a tunnel accident, namely, in a certain time, M business frequency is more than or equal to a new security risk table, the M business data is calculated, the security risk table is further, the security risk level is triggered by the security risk table is calculated, the security risk level is further, the security risk level is calculated, the security risk level is more than the security risk level is triggered by the security risk level is calculated, and the security risk level is more than the security risk level is calculated by the security risk level is calculated, the method comprises the steps of matching the security risk coefficient in a security risk level table as index data, acquiring the security risk level of the tunnel according to a matching result, wherein the security risk coefficient and the security risk level in the security risk level table are in one-to-one correspondence, so that the security risk coefficient and the security risk level are used as reference data when defect detection is carried out for the later digital twin-based tunnel construction.
Step S600: and when the security risk level is greater than or equal to a security risk level threshold, generating dangerous identification information for the image abnormal characteristic region, synchronizing the dangerous identification information to the tunnel digital twin model, and sending the dangerous identification information to a user side.
Specifically, the image abnormal characteristic region and the region abnormal coefficient are sent to a software processing console for security risk association, the obtained security risk level is judged, when the security risk level is greater than or equal to a security risk level threshold, the image abnormal characteristic region is generated with risk identification information, the security risk level threshold is a standard for formulating the security risk level according to the industry security standard during construction in a tunnel, and a reasonable critical value is set. When the obtained security risk level is greater than or equal to the security risk level, relief measures and emergency measures are needed to be taken, namely dangerous identification is carried out on the image abnormal characteristic area through running of the tag, the risk in the tunnel is realized within a controllable range, the image abnormal characteristic area data with dangerous identification information are synchronized into the constructed tunnel digital twin model, the same tag identification information is also arranged in the tunnel digital twin model, and finally the tunnel digital twin model with the dangerous identification tag is sent to a user side, so that reminding of a dangerous area of a user in the tunnel is realized, and the accuracy of defect detection based on digital twin tunnel construction is improved in the later period.
In summary, the method for detecting the tunnel construction defect based on the digital twinning at least includes the following technical effects that the method for detecting the tunnel construction defect at least includes the steps that first tunnel region positioning information is received, a first tunnel region reference image is downloaded, image capturing is conducted to generate a first tunnel region monitoring image, the first tunnel region reference image and the first tunnel region monitoring image are received, an image abnormal feature region and a region abnormal coefficient are generated, a security risk level is obtained, when the security risk level is greater than or equal to a security risk level threshold value, danger identification information is generated for the image abnormal feature region and is synchronized to a tunnel digital twinning model, the tunnel digital twinning model is sent to a user side, the reference image with high scene combination degree is obtained through the digital twinning model, and defect monitoring with high accuracy is achieved.
Embodiment two:
based on the same inventive concept as the method for detecting a tunnel construction defect based on digital twinning in the foregoing embodiment, as shown in fig. 5, the present application provides a system for detecting a tunnel construction defect based on digital twinning, the system comprising:
the information receiving module 1 is used for receiving first tunnel region positioning information when the tunnel construction defect identification robot runs to a first tunnel section, wherein the first tunnel region belongs to the first tunnel section;
The reference image module 2 is used for downloading a first tunnel region reference image from a tunnel digital twin model of the software processing platform according to the first tunnel region positioning information;
the image capturing module 3 is used for acquiring shooting action characteristics of the first tunnel region reference image, controlling a vision module of the tunnel construction defect recognition robot to capture images, and generating a first tunnel region monitoring image;
the abnormality module 4 is used for activating a first twin node embedded in an edge processor of the tunnel construction defect identification robot to receive the first tunnel region reference image, and activating a second twin node of the edge processor to receive the first tunnel region monitoring image, so as to generate an image abnormal characteristic region and a region abnormality coefficient;
the security risk association module 5 is configured to send the image abnormal feature region and the region abnormal coefficient to the software processing console for security risk association when the region abnormal coefficient is greater than or equal to an abnormal coefficient critical value, and obtain a security risk level;
The first judging module 6 is configured to generate dangerous identification information for the image abnormal feature area when the security risk level is greater than or equal to a security risk level threshold, synchronize the dangerous identification information with the tunnel digital twin model, and send the dangerous identification information to a user terminal.
Further, the system further comprises:
the model construction module is used for uploading construction design original pictures and quality design standards of a preset tunnel through a user side to construct the digital twin model of the preset tunnel;
the positioning module is used for positioning the preset tunnel in a grid space coordinate system and synchronizing the preset tunnel to the tunnel digital twin model;
the state conversion module is used for acquiring positioning information of the identification robot according to the Beidou positioning module of the tunnel construction defect identification robot, synchronizing the positioning information of the identification robot to the tunnel digital twin model, and converting the robot into a static state when the tunnel construction defect identification robot runs to the first tunnel section to acquire the positioning information of the first tunnel region.
Further, the system further comprises:
the first activation module is used for activating a photosensitive module of the tunnel construction defect identification robot and acquiring sensing illumination vector information, wherein the illumination vector information comprises light intensity sensing information and illumination direction sensing information;
A first synchronization module for synchronizing the light intensity sensing information and the illumination direction sensing information to the tunnel digital twin model based on the first tunnel region positioning information, generating the first tunnel region reference image, comprising,
when the first tunnel region positioning information is received, a first tunnel region initial image is acquired, wherein the first tunnel region initial image has initial illumination vector information;
the encoding module is used for activating an image encoder embedded in the tunnel digital twin model, encoding the initial image of the first tunnel region based on the initial illumination vector information and the sensing illumination vector information, and generating the reference image of the first tunnel region.
Further, the system further comprises:
the identification module is used for acquiring a first tunnel region identification image, and the first tunnel region identification image comprises a sensing illumination record vector;
the mapping module is used for mapping the initial illumination vector information, the sensing illumination record vector and the first tunnel region initial image through a convolution topological network to generate a first tunnel region coded image;
The similarity evaluation module is used for performing similarity evaluation on the first tunnel region coded image and the first tunnel region identification image to obtain a mapping loss amount, wherein the mapping loss amount is a similarity negative value;
and the second judging module is used for generating the image encoder when the continuous preset times of the mapping loss amount are smaller than the mapping preset loss amount.
Further, the system further comprises:
the grouping module is used for grouping the first tunnel region monitoring image and the first tunnel region reference image according to the shooting action characteristics, and obtaining a plurality of groups of first tunnel region monitoring images and first tunnel region reference images with consistent shooting action characteristics;
the first morphological feature extraction module is used for extracting morphological features of the first tunnel region reference image by utilizing the first twin node to obtain first morphological feature information;
the second morphological feature extraction module is used for extracting morphological features of the first tunnel region monitoring image by utilizing the second twin nodes to obtain second morphological feature information;
The deviation calculation module is used for carrying out deviation calculation on the first morphological feature information and the second morphological feature information to obtain the image abnormal feature region with the feature deviation distance being greater than or equal to a deviation distance threshold;
and the average module is used for setting the average value of the characteristic deviation distance of the image abnormal characteristic region as the region abnormal coefficient.
Further, the system further comprises:
the first Euclidean distance calculation module is used for obtaining the first morphological feature information and the second morphological feature information of a first pixel point, performing Euclidean distance calculation, and obtaining a first feature deviation distance;
the second Euclidean distance calculation module is used for obtaining the first morphological feature information and the second morphological feature information of the N-th pixel point, carrying out Euclidean distance calculation and obtaining the N-th feature deviation distance;
the traversing module is used for traversing the first characteristic deviation distance to the Nth characteristic deviation distance and adding pixel points which are larger than or equal to the deviation distance threshold value into an abnormal pixel point set;
The partitioning module is used for partitioning by traversing the pixel point position information of the abnormal pixel point set according to the region continuous distance threshold value, and generating the image abnormal characteristic region.
Further, the system further comprises:
the retrieval module is used for retrieving in the tunnel engineering big data according to the tunnel digital twin model, the image abnormal characteristic area and the area abnormal coefficient to generate M pieces of construction transaction data, wherein the M pieces of construction transaction data comprise accident triggering frequency, M is more than or equal to 500, and M is an integer;
the first calculation module is used for calculating the safety risk coefficient, wherein the safety risk coefficient is the ratio of the accident triggering frequency to the M pieces of construction business data;
the second activation module is used for activating a security risk level table to process the security risk coefficient and obtain the security risk level, wherein the security risk coefficient and the security risk level in the security risk level table are in one-to-one correspondence.
In the foregoing description of a method for detecting a defect in a tunnel construction based on digital twin, it will be clear to those skilled in the art that a system for detecting a defect in a tunnel construction based on digital twin in this embodiment is relatively simple in description, and relevant places refer to the description of the method section, because the device disclosed in the embodiment corresponds to the method disclosed in the embodiment.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. The method is characterized by being applied to a digital twinning-based tunnel construction defect detection system, wherein the system comprises a software processing console and a hardware execution terminal, the hardware execution terminal comprises a tunnel construction defect identification robot and a user side, and the method comprises the following steps:
when a tunnel construction defect identification robot runs to a first tunnel section, receiving first tunnel region positioning information, wherein a first tunnel region belongs to the first tunnel section;
according to the first tunnel region positioning information, downloading a first tunnel region reference image from a tunnel digital twin model of a software processing console;
Acquiring shooting action characteristics of the first tunnel region reference image, and controlling a vision module of the tunnel construction defect recognition robot to perform image capturing so as to generate a first tunnel region monitoring image;
activating a first twin node embedded in an edge processor of the tunnel construction defect identification robot to receive the first tunnel region reference image, and activating a second twin node of the edge processor to receive the first tunnel region monitoring image to generate an image abnormal characteristic region and a region abnormal coefficient;
when the regional abnormal coefficient is larger than or equal to an abnormal coefficient critical value, the image abnormal characteristic region and the regional abnormal coefficient are sent to the software processing console for security risk association, and a security risk level is obtained;
and when the security risk level is greater than or equal to a security risk level threshold, generating dangerous identification information for the image abnormal characteristic region, synchronizing the dangerous identification information to the tunnel digital twin model, and sending the dangerous identification information to a user side.
2. The method of claim 1, wherein receiving first tunnel region location information when the tunnel construction defect identification robot travels to a first tunnel section, wherein the first tunnel region belongs to the first tunnel section, comprises:
Uploading a construction design original image and a quality design standard of a preset tunnel by a user side, and constructing the tunnel digital twin model of the preset tunnel;
positioning the preset tunnel in a grid space coordinate system, and synchronizing to the tunnel digital twin model;
and according to the Beidou positioning module of the tunnel construction defect identification robot, acquiring positioning information of the identification robot, synchronizing the positioning information to the tunnel digital twin model, and converting the robot into a static state when the tunnel construction defect identification robot runs to the first tunnel section to acquire the positioning information of the first tunnel region.
3. The method of claim 2, wherein downloading the first tunnel region reference image from the tunnel digital twin model of the software processing console based on the first tunnel region location information comprises:
activating a photosensitive module of the tunnel construction defect identification robot to acquire sensing illumination vector information, wherein the illumination vector information comprises light intensity sensing information and illumination direction sensing information;
synchronizing the light intensity sensing information and the light direction sensing information to the tunnel digital twin model based on the first tunnel region positioning information, generating the first tunnel region reference image, including,
When the first tunnel region positioning information is received, a first tunnel region initial image is acquired, wherein the first tunnel region initial image has initial illumination vector information;
activating an image encoder embedded in the tunnel digital twin model, encoding the initial image of the first tunnel region based on the initial illumination vector information and the sensing illumination vector information, and generating the reference image of the first tunnel region.
4. The method of claim 3, wherein activating an image encoder embedded in the tunnel digital twin model, encoding the first tunnel region initial image based on the initial illumination vector information and the sensed illumination vector information, generating the first tunnel region reference image comprises:
acquiring a first tunnel region identification image, wherein the first tunnel region identification image comprises a sensing illumination record vector;
mapping the initial illumination vector information, the sensing illumination record vector and the first tunnel region initial image through a convolution topological network to generate a first tunnel region coded image;
performing similarity evaluation on the first tunnel region coding image and the first tunnel region identification image to obtain a mapping loss amount, wherein the mapping loss amount is a similarity negative value;
And when the mapping loss amount is continuously preset for times smaller than the mapping preset loss amount, generating the image encoder.
5. The method of claim 1, wherein activating a first twin node embedded in an edge processor of the tunnel construction defect identification robot to receive the first tunnel region reference image and activating a second twin node of the edge processor to receive the first tunnel region monitoring image generates an image anomaly characteristic region and a region anomaly coefficient, comprises:
according to the shooting action characteristics, respectively grouping the first tunnel region monitoring image and the first tunnel region reference image to obtain a plurality of groups of first tunnel region monitoring images and first tunnel region reference images with consistent shooting action characteristics;
performing morphological feature extraction on the first tunnel region reference image by using the first twin node to obtain first morphological feature information;
performing morphological feature extraction on the first tunnel region monitoring image by using the second twin node to obtain second morphological feature information;
performing deviation calculation on the first morphological feature information and the second morphological feature information to obtain the image abnormal feature region with the feature deviation distance being greater than or equal to a deviation distance threshold;
And setting the average value of the characteristic deviation distances of the image abnormal characteristic areas as the area abnormal coefficient.
6. The method of claim 5, wherein performing a deviation calculation on the first topographical feature information and the second topographical feature information to obtain the image anomaly feature area having a feature deviation distance greater than or equal to a deviation distance threshold comprises:
acquiring the first morphological feature information and the second morphological feature information of a first pixel point, and performing Euclidean distance calculation to acquire a first feature deviation distance;
acquiring the first morphological feature information and the second morphological feature information of the N pixel point, and performing Euclidean distance calculation to acquire an N feature deviation distance;
traversing the first characteristic deviation distance to the Nth characteristic deviation distance, and adding pixel points which are larger than or equal to the deviation distance threshold value into an abnormal pixel point set;
and traversing the pixel point position information of the abnormal pixel point set to partition according to the region continuous distance threshold value, and generating the image abnormal characteristic region.
7. The method of claim 1, wherein when the regional anomaly coefficient is greater than or equal to an anomaly coefficient threshold value, sending the image anomaly characteristic region and the regional anomaly coefficient to the software processing console for security risk association, obtaining a security risk level, comprising:
Searching in the tunnel engineering big data according to the tunnel digital twin model, the image abnormal characteristic region and the region abnormal coefficient to generate M pieces of construction transaction data, wherein the M pieces of construction transaction data comprise accident triggering frequency, M is more than or equal to 500, and M is an integer;
calculating a safety risk coefficient, wherein the safety risk coefficient is the ratio of the accident triggering frequency to the M pieces of construction business data;
and activating a security risk level table to process the security risk coefficient to acquire the security risk level, wherein the security risk coefficient and the security risk level in the security risk level table are in one-to-one correspondence.
8. The utility model provides a tunnel construction defect detecting system based on digit twin which characterized in that, the system includes software processing center and hardware execution terminal, the hardware execution terminal includes tunnel construction defect identification robot and user side, includes:
the information receiving module is used for receiving positioning information of a first tunnel area when the tunnel construction defect identification robot runs to the first tunnel section, wherein the first tunnel area belongs to the first tunnel section;
The reference image module is used for downloading a first tunnel region reference image from a tunnel digital twin model of the software processing console according to the first tunnel region positioning information;
the image capturing module is used for acquiring shooting action characteristics of the first tunnel region reference image, controlling a vision module of the tunnel construction defect recognition robot to capture images and generating a first tunnel region monitoring image;
the abnormality module is used for activating a first twin node embedded in an edge processor of the tunnel construction defect identification robot to receive the first tunnel region reference image, and activating a second twin node of the edge processor to receive the first tunnel region monitoring image so as to generate an image abnormality characteristic region and a region abnormality coefficient;
the security risk association module is used for sending the image abnormal characteristic region and the region abnormal coefficient to the software processing platform for security risk association when the region abnormal coefficient is greater than or equal to an abnormal coefficient critical value, and obtaining a security risk level;
the first judging module is used for generating dangerous identification information for the image abnormal characteristic area when the security risk level is greater than or equal to a security risk level threshold value, synchronizing the dangerous identification information to the tunnel digital twin model and sending the dangerous identification information to a user side.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311181674.1A CN116908214B (en) | 2023-09-14 | 2023-09-14 | Tunnel construction defect detection method and system based on digital twinning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311181674.1A CN116908214B (en) | 2023-09-14 | 2023-09-14 | Tunnel construction defect detection method and system based on digital twinning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116908214A true CN116908214A (en) | 2023-10-20 |
| CN116908214B CN116908214B (en) | 2023-12-08 |
Family
ID=88353490
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311181674.1A Active CN116908214B (en) | 2023-09-14 | 2023-09-14 | Tunnel construction defect detection method and system based on digital twinning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116908214B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117549304A (en) * | 2023-12-06 | 2024-02-13 | 中交三公局第六工程(河北)有限公司 | Digital twinning-based robot tunnel blasting pre-detection control method and system |
| CN117671303A (en) * | 2024-02-02 | 2024-03-08 | 江苏濠汉信息技术有限公司 | Stereoscopic ranging hidden danger early warning method and system based on twin network |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113822227A (en) * | 2021-10-19 | 2021-12-21 | 杭州交通工程监理咨询有限公司 | Tunnel engineering construction risk monitoring method, device and medium |
| US20220167142A1 (en) * | 2020-11-24 | 2022-05-26 | Electronics And Telecommunications Research Institute | Digital twin disaster management system customized for underground public areas |
| CN116070407A (en) * | 2022-11-28 | 2023-05-05 | 华中科技大学 | Digital twin optimization method and system for open cut construction of upward-penetrating existing subway |
| CN116402866A (en) * | 2023-03-03 | 2023-07-07 | 西安交通大学 | Point cloud-based part digital twin geometric modeling and error assessment method and system |
| CN116680786A (en) * | 2023-06-02 | 2023-09-01 | 广东电网有限责任公司广州供电局 | Digital twin modeling system suitable for operation and maintenance of power tunnel |
| CN116720242A (en) * | 2023-05-23 | 2023-09-08 | 北京国网富达科技发展有限责任公司 | Digital twin panoramic monitoring system for high-voltage cable tunnel |
-
2023
- 2023-09-14 CN CN202311181674.1A patent/CN116908214B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20220167142A1 (en) * | 2020-11-24 | 2022-05-26 | Electronics And Telecommunications Research Institute | Digital twin disaster management system customized for underground public areas |
| CN113822227A (en) * | 2021-10-19 | 2021-12-21 | 杭州交通工程监理咨询有限公司 | Tunnel engineering construction risk monitoring method, device and medium |
| CN116070407A (en) * | 2022-11-28 | 2023-05-05 | 华中科技大学 | Digital twin optimization method and system for open cut construction of upward-penetrating existing subway |
| CN116402866A (en) * | 2023-03-03 | 2023-07-07 | 西安交通大学 | Point cloud-based part digital twin geometric modeling and error assessment method and system |
| CN116720242A (en) * | 2023-05-23 | 2023-09-08 | 北京国网富达科技发展有限责任公司 | Digital twin panoramic monitoring system for high-voltage cable tunnel |
| CN116680786A (en) * | 2023-06-02 | 2023-09-01 | 广东电网有限责任公司广州供电局 | Digital twin modeling system suitable for operation and maintenance of power tunnel |
Non-Patent Citations (1)
| Title |
|---|
| 梁亚成等: "基于数字孪生的隧道智能巡检技术研究与应用", 《北京测绘》, vol. 36, no. 07, pages 870 - 874 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117549304A (en) * | 2023-12-06 | 2024-02-13 | 中交三公局第六工程(河北)有限公司 | Digital twinning-based robot tunnel blasting pre-detection control method and system |
| CN117549304B (en) * | 2023-12-06 | 2024-05-07 | 中交三公局第六工程(河北)有限公司 | Digital twinning-based robot tunnel blasting pre-detection control method and system |
| CN117671303A (en) * | 2024-02-02 | 2024-03-08 | 江苏濠汉信息技术有限公司 | Stereoscopic ranging hidden danger early warning method and system based on twin network |
| CN117671303B (en) * | 2024-02-02 | 2024-04-12 | 江苏濠汉信息技术有限公司 | Stereoscopic ranging hidden danger early warning method and system based on twin network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116908214B (en) | 2023-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116908214B (en) | Tunnel construction defect detection method and system based on digital twinning | |
| US20210319561A1 (en) | Image segmentation method and system for pavement disease based on deep learning | |
| CN105051754B (en) | Method and apparatus for detecting people by monitoring system | |
| CN105158257B (en) | Slide plate measurement method and device | |
| CN109448326B (en) | Geological disaster intelligent group defense monitoring system based on rapid image recognition | |
| CN107977656A (en) | A kind of pedestrian recognition methods and system again | |
| EP3023912A1 (en) | Crack data collection apparatus and server apparatus to collect crack data | |
| US6141440A (en) | Disparity measurement with variably sized interrogation regions | |
| KR102522928B1 (en) | Method and apparatus for estimating moving velocity of vehicle | |
| CN115546116B (en) | Full-coverage type rock mass discontinuous surface extraction and interval calculation method and system | |
| CN112541403B (en) | Indoor personnel falling detection method by utilizing infrared camera | |
| KR102559586B1 (en) | Structural appearance inspection system and method using artificial intelligence | |
| WO2020093631A1 (en) | Antenna downtilt angle measurement method based on depth instance segmentation network | |
| CN112652021A (en) | Camera offset detection method and device, electronic equipment and storage medium | |
| CN115639248A (en) | System and method for detecting quality of building outer wall | |
| JP3701167B2 (en) | Water level measuring method and apparatus | |
| CN115035182A (en) | Landslide disaster early warning method and system in mountainous area | |
| CN105043355B (en) | Side slope Light deformation monitoring method and monitoring system based on similarity decision criteria | |
| CN115512247A (en) | Regional building damage grade assessment method based on image multi-parameter extraction | |
| CN109086350A (en) | A kind of mixed image search method based on WiFi | |
| CN110636248B (en) | Target tracking method and device | |
| Sowmyayani et al. | Fall detection in elderly care system based on group of pictures | |
| WO2024067435A1 (en) | Video-based multi-object displacement tracking monitoring method and apparatus | |
| CN104504162B (en) | A kind of video retrieval method based on robot vision platform | |
| CN115880643B (en) | Social distance monitoring method and device based on target detection algorithm |
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 |