CN112329705B - Method for monitoring maximum deformation in lifting process of large-scale structure - Google Patents

Method for monitoring maximum deformation in lifting process of large-scale structure Download PDF

Info

Publication number
CN112329705B
CN112329705B CN202011316723.4A CN202011316723A CN112329705B CN 112329705 B CN112329705 B CN 112329705B CN 202011316723 A CN202011316723 A CN 202011316723A CN 112329705 B CN112329705 B CN 112329705B
Authority
CN
China
Prior art keywords
point
center point
cluster
coordinates
central point
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
Application number
CN202011316723.4A
Other languages
Chinese (zh)
Other versions
CN112329705A (en
Inventor
孙舸
才立彬
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Bomesc Offshore Engineering Co Ltd
Original Assignee
Bomesc Offshore Engineering Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Bomesc Offshore Engineering Co Ltd filed Critical Bomesc Offshore Engineering Co Ltd
Priority to CN202011316723.4A priority Critical patent/CN112329705B/en
Publication of CN112329705A publication Critical patent/CN112329705A/en
Application granted granted Critical
Publication of CN112329705B publication Critical patent/CN112329705B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/21—Design, administration or maintenance of databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/22—Indexing; Data structures therefor; Storage structures
    • G06F16/2282—Tablespace storage structures; Management thereof
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Multimedia (AREA)
  • Probability & Statistics with Applications (AREA)
  • Software Systems (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

The invention discloses a method for monitoring the maximum deformation in the lifting process of a large-scale structure, which comprises the steps of carrying out laser irradiation on the surface of the structure to form laser speckles, using a CCD camera to shoot according to a certain time interval when the structure is not lifted and in the lifting process, processing a gray-scale image obtained by shooting by using an image acquisition card to obtain pixel point coordinates of the speckles, carrying out clustering processing on coordinate data by using a K-means module in Python, selecting a clustering central point in each speckle as an approximate substitute of the position of the speckle, calculating the offset of corresponding pixel points in matlab, obtaining the maximum offset, drawing a line graph of which the offset changes along with time, and immediately stopping the machine for maintenance if the line graph has special conditions in the lifting process. The invention can improve the measurement precision of the surface deformation of the structure, simplify the setting process of the measuring device, reduce the use cost and greatly improve the measurement efficiency.

Description

Method for monitoring maximum deformation amount in large-scale structure lifting process
Technical Field
The invention relates to a method for monitoring deformation of a large structure, in particular to a method for monitoring maximum deformation of the large structure in a lifting process.
Background
When a large-sized structure is installed and transported, the large-sized structure needs to be lifted through a lifting point on the structure or a fulcrum on the structure is lifted, the large-sized structure can be deformed under the unstable condition generated in the lifting or lifting process, and engineering accidents can be caused when the deformation of the large-sized structure exceeds a critical value.
The span of the large-scale structure is large, the allowable deformation in the lifting process is small, and the structure deformation in the lifting process of the large-scale structure is difficult to accurately measure by using a traditional displacement sensor and an angle inclination sensor. The more accurate measurement method is to measure the surface to be measured by adopting a static level gauge, but the method has complex device setting process, usually needs debugging for several days for one-time measurement, can only measure aiming at a horizontal plane, and has less application scenes.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a non-contact real-time monitoring method which can improve the deformation measurement precision, simplify the device setting process, reduce the use cost and greatly improve the measurement efficiency.
The invention discloses a method for monitoring the maximum deformation in the lifting process of a large-scale structure, which comprises the following steps:
step one, collecting a gray level image of an initial state when the surface of a structure is not lifted and a state in the lifting process by using a CCD camera, wherein the specific process is as follows:
the method comprises the following steps that firstly, laser is adopted to irradiate the surface of a structure, so that a plurality of laser speckles which are approximately circular are generated on the surface of the structure, the central lines of the laser speckles are parallel to each other and are uniformly distributed on the surface of the structure, and the position of a CCD camera is adjusted, so that the lens plane of the CCD camera is parallel to the surface of the structure which generates the laser speckles;
secondly, shooting the surface of the structure generating the laser speckles by using a CCD camera according to a set time interval when the structure is not lifted and in the lifting process, and obtaining a gray level image of the laser speckles on the surface of the structure at different moments;
and secondly, performing data extraction and data storage on each gray-scale image obtained by the CCD camera by using SQL Server software, wherein the specific process is as follows:
firstly, SQL Server software is used for creating an image information database, and then n laser speckle pixel point position tables A are established in the image information databaseiN cluster central point disorder table UiAnd n cluster central points ordered table SiI is 1 … n, and the tables with the same subscript i have corresponding relation with each other;
secondly, transmitting the gray level image obtained by the CCD camera to a computer externally connected with an image acquisition card through a network or a circuit;
thirdly, the computer reads the gray level images of the laser speckles on the surface of the structure at different moments by using an image acquisition card, establishes an image coordinate system with the coordinate origin positioned at the lower left corner of the image in each gray level image, records the total number k of the laser speckles, selects a pixel point in each laser speckle image according to a set sequence as an approximate substitution position of the laser speckle position and extracts selected pixel point coordinate data, and sequentially and respectively stores each pixel point coordinate data selected from each gray level image into a laser speckle pixel point position table corresponding to each gray level image in a list form;
step three, utilizing a K-means module in Python to respectively perform a table A of the positions of the laser speckle pixel pointsiCarrying out clustering processing on coordinate data of each pixel point, wherein the specific process is as follows:
firstly, leading in a K-means module in Python, establishing connection with an SQL Server database, and extracting a laser speckle pixel point position table Ai；
Secondly, the position table A of each laser speckle pixel point is usediDividing all pixel point coordinate data into K feature spaces with similar sizes in sequence from front to back in a list by using a K-means module respectively, and randomly selecting a point in each feature space as an initial clustering central point;
thirdly, respectively calculating a position table A of each laser speckle pixel pointiThe distance from all pixel points to each initial clustering center point is compared, then the distance from each pixel point to each initial clustering center point is compared, each pixel point closest to each initial clustering center point is recombined with the initial clustering center point to form a reconstruction feature space, k reconstruction feature spaces are formed,
fourthly, calculating a position table A of each laser speckle pixel pointiMean value of horizontal and vertical coordinates of all pixel points in each reconstruction characteristic space
Figure GDA0003568086610000031
And
Figure GDA0003568086610000032
formed point
Figure GDA0003568086610000033
) As the coordinates of the new cluster center point of each reconstructed feature space;
fifthly, adopting a position table A of each laser speckle pixel pointiThe coordinates of the new clustering center point of each reconstructed feature space replace the coordinates of the initial clustering center point in the third step, and the third step and the fifth step are repeated until the laser speckle pixel point position table AiThe coordinates of the new clustering center point in each reconstructed feature space are not changed any more, and the coordinates of the final clustering center point are obtained;
sixthly, the position table A of each laser speckle pixel point in the fifth step is displayediThe coordinates of each final clustering central point in the cluster central point list are respectively stored into a clustering central point disorder table U in an SQL database corresponding to the laser speckle pixel point position tableiPerforming the following steps;
step four, using SQL Server software to carry out data sequencing and index correspondence to the cluster center point unordered table, and the concrete process is as follows:
firstly, SQL Server software is adopted to arrange a table S in order at each cluster central pointiAllocating k blank storage areas in the middle, and sequentially allocating 1 … k index values;
step two, respectively calculating the ith cluster central point disorder table UiIn each final clustering center point and the (i-1) th clustering center point ordered table Si-1The distance between the k coordinate points is sorted with the i-1 st cluster center pointi-1The ith cluster center point disorder table U with each coordinate point nearest to the coordinate pointiThe final clustering center point coordinate in the table S is sorted in the ith-1 clustering center point according to the final clustering center pointi-1The index value in the table is stored in the same index value position in the ith cluster central point ordered table; wherein i-0 indicates that the structure is not lifted, and in the 0 th clusterCentral point disorder table U0The coordinates of the final cluster central point in the table are directly and sequentially stored into the 0 th cluster central point ordered table S0Performing the following steps;
step five, drawing the image of the offset of the deformation area on the surface of the structure by utilizing matlab software, wherein the specific process is as follows:
step one, using the 0 th clustering central point to sequence the table S0Taking the coordinates of each final clustering center point as a reference, and calculating the 0 th clustering center point ordered list S by using matlab software0The coordinates of each final clustering center point in the table are respectively compared with the ith clustering center point ordered table SiThe position offset epsilon between the coordinates of each cluster central point at the position with the same middle index valuejAnd an amount of angular offset
Figure GDA0003568086610000041
Secondly, obtaining each cluster central point ordered list S by utilizing the math toolkit in the matlabiEach final cluster central point and 0 th cluster central point ordered table S0The maximum position offset epsilon of the structure surface is compared with the final clustering center point of each same index valuemaxAnd an amount of angular offset
Figure GDA0003568086610000042
Then, the data is sorted by the cluster central point SiThe shooting time corresponding to the gray-scale map of the laser speckle of the surface of the structure respectively corresponds to X-axis coordinates, and the maximum position offset epsilon of the surface of the structure corresponding to each cluster central point ordered tablemaxAnd an amount of angular offset
Figure GDA0003568086610000043
Drawing a line graph of the change of the offset with time for the ordinate;
and step six, in the line graph of the deviation amount changing along with the time in the step five, if the maximum position deviation amount and the maximum angle deviation amount of the surface of the structure are close to the preset critical deformation amount, the curve slope is overlarge and the curve slope generates a sudden change, the lifting work is immediately stopped, the related fixed lifting point is fixedly strengthened, and the lifting is continued after the lifting standard is reached.
The invention has the advantages that: the method has the advantages that the deformation of each point on the large-scale structure in the lifting process is measured in real time by using a high-precision image recognition technology, the deformation is monitored in real time, engineering accidents caused by the fact that the deformation of the large-scale structure exceeds a critical value are prevented, the method can be used for solving the problem of measuring the maximum deformation of the large-scale structure in the installation and transportation processes of the large-scale structure, and the method is simple in operation, visible in result, stable, reliable and wide in adaptability.
Drawings
FIG. 1 is a flow chart of a method for monitoring maximum deformation during the lifting process of a large structure.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments.
The working principle of the invention is as follows: 1. firstly, carrying out laser irradiation on the surface of a structure to form approximately circular speckles, wherein all rows and all columns of the speckles are parallel to each other, and the speckles are similar in size and uniform in distribution; 2. in the lifting process, the speckles can be changed in position and shape, but because the deformation of the large-scale structure belongs to micro deformation, each row and each column of the speckles can still be regarded as approximately parallel in the lifting process; 3. selecting a small pixel point in each speckle as an approximate substitute of the speckle position; 4. calculating the offset of the corresponding pixel points can approximately replace the offset of the laser speckles, and the deformation of the surface of the structure is monitored through the offset of the laser speckles.
Based on the principle, the method of the invention comprises the following steps:
as shown in fig. 1, the method for monitoring the maximum deformation amount in the lifting process of the large structure of the present invention comprises the following steps:
step one, collecting a gray level image of an initial state when the surface of a structure is not lifted and a state in the lifting process by using a CCD camera, wherein the specific process is as follows:
the method comprises the steps that firstly, the surface of a structure is irradiated by laser to enable the surface of the structure to generate a plurality of laser speckles which are approximately circular, the central lines of the laser speckles are parallel to each other and evenly distributed on the surface of the structure, the position of a CCD camera is adjusted to enable the lens plane of the CCD camera to be parallel to the surface of the structure which generates the laser speckles, and measuring errors are reduced.
And secondly, shooting the surface of the structure generating the laser speckles by using a CCD camera according to a set time interval (such as 0.1S) when the structure is not lifted and in the lifting process, and obtaining a gray level image of the laser speckles on the surface of the structure at different moments.
And secondly, performing data extraction and data storage on each gray-scale image obtained by the CCD camera by using SQL Server software, wherein the specific process is as follows:
firstly, SQL Server software is used for creating an image information database, and then n laser speckle pixel point position tables A are established in the image information databaseiN cluster central point disorder table UiAnd n cluster central points ordered table SiI is 1 … n, and the tables with the same index i have a corresponding relationship with each other.
And secondly, transmitting the gray level image obtained by the CCD camera to a computer externally connected with an image acquisition card through a network or a circuit.
Thirdly, the computer reads the gray level images of the laser speckles on the surface of the structure at different moments by using an image acquisition card, establishes an image coordinate system with the coordinate origin positioned at the lower left corner of the image in each gray level image, records the total number k of the laser speckles, selects a pixel point in each laser speckle image as an approximate substitution position of the laser speckle position according to a set sequence (from top to bottom and from left to right, for example), extracts the coordinate data of the selected pixel point, and sequentially and respectively stores the coordinate data of each pixel point selected from each gray level image into a laser speckle pixel point position table corresponding to each gray level image in a list form, namely the coordinate of the pixel point selected from the same gray level image is stored into the same laser speckle pixel point position table.
Thirdly, respectively carrying out speckle image processing on each laser by using a K-means module in PythonPlain dot position table AiClustering the coordinate data of each pixel point, wherein the specific process is as follows:
firstly, leading in a K-means module in Python, establishing connection with an SQL Server database, and extracting a laser speckle pixel point position table Ai。
Secondly, the position table A of each laser speckle pixel point is usediAll the pixel point coordinate data in the cluster are divided into K feature spaces with similar sizes in sequence from front to back in the list by using a K-means module respectively, and a point is randomly selected in each feature space to serve as an initial clustering central point.
Thirdly, respectively calculating a position table A of each laser speckle pixel pointiThe distance from all the internal pixel points to each initial clustering center point is compared, then the distance from each pixel point to each initial clustering center point is compared, each pixel point closest to each initial clustering center point is recombined with the initial clustering center point to form a reconstruction feature space, k reconstruction feature spaces are formed, and the distance formula is as follows:
Figure GDA0003568086610000061
in the formula
Figure GDA0003568086610000062
And
Figure GDA0003568086610000063
table A for indicating any laser speckle pixel positioniThe abscissa and ordinate of any one of the pixel points,
Figure GDA0003568086610000064
and
Figure GDA0003568086610000065
table A for indicating any laser speckle pixel positioniThe abscissa and the ordinate of the initial clustering center point corresponding to any one feature space.
Fourthly, calculating a position table A of each laser speckle pixel pointiMean value of horizontal and vertical coordinates of all pixel points in each reconstruction characteristic space
Figure GDA0003568086610000066
And
Figure GDA0003568086610000067
formed point
Figure GDA0003568086610000068
) As the coordinates of the new cluster center point for each reconstructed feature space, wherein:
Figure GDA0003568086610000069
Figure GDA00035680866100000610
in the formula
Figure GDA0003568086610000071
Respectively representing the abscissa sum and the ordinate sum, n, of all pixel points in each reconstruction spacejAnd representing the total number of all pixel points in each reconstruction space. j denotes the jth feature space, njRepresenting the total number of pixel points in the jth characteristic space, w representing the w-th pixel point in the jth characteristic space,
Figure GDA0003568086610000072
and
Figure GDA0003568086610000073
and (4) representing the horizontal and vertical coordinates of the w-th pixel point in the j-th feature space.
Fifthly, adopting a position table A of each laser speckle pixel pointiCoordinates of the new cluster center point of each reconstructed feature space in the third step replace the coordinates of the initial cluster center point in the third stepAnd repeating the third step and the fifth step until the laser speckle pixel point position table AiThe coordinates of the new clustering center point in each reconstructed feature space are not changed any more, and the coordinates of the final clustering center point are obtained.
Sixthly, the position table A of each laser speckle pixel point in the fifth step is displayediThe coordinates of each final clustering central point in the cluster central point list are respectively stored into a clustering central point disorder table U in an SQL database corresponding to the laser speckle pixel point position tableiIn (1).
Step four, using SQL Server software to carry out data sequencing and index correspondence to the cluster center point unordered table, and the specific process is as follows:
firstly, SQL Server software is adopted to arrange a table S in each cluster central pointiAllocating k blank storage areas in the middle, and sequentially allocating 1 … k index values;
step two, respectively calculating the ith cluster central point disorder table UiIn each final clustering center point and the (i-1) th clustering center point ordered table Si-1The distance between the k coordinate points is sorted with the i-1 st cluster center pointi-1In the ith cluster central point disorder table U with each coordinate point closest to each otheriThe final clustering center point coordinate in the table S is sorted in the ith-1 clustering center point according to the final clustering center pointi-1The index value in the table is stored in the same index value position in the ith cluster central point ordered table; wherein, i is 0 and represents the structure is not lifted, the 0 th cluster central point unordered table U0The coordinates of the final cluster central point in the table are directly and sequentially stored into the 0 th cluster central point ordered table S0In (1).
This step is illustrated below by way of example: suppose that the 6 th cluster center point unordered table U is currently being sorted6Processing is carried out, and the 5 th cluster central point ordered table S5After completion of the filling, 3 pieces of coordinate data are stored in the two tables, respectively. Firstly, in the 6 th cluster center point ordered table S63 blank storage areas are distributed and index values of 1-3 are distributed in sequence; then, take out U61 st coordinate d1Separately calculate d1And S5Distance of 3 coordinates in, let d1Closest to the 2 nd coordinate point, the index value 2 is obtained, and S is carried out6Find the blank storage area with index 2, and find d1Storing the position; then, for U6Middle 2 nd coordinate d23 rd coordinate d3Performing the same operation, and adding S6And (5) completing filling.
Step five, drawing the image of the offset of the deformation area on the surface of the structure by utilizing matlab software, wherein the specific process is as follows:
step one, using the 0 th clustering central point to sequence the table S0Taking the coordinates of each final clustering central point as a reference, and adopting matlab software to calculate the 0 th clustering central point ordered list S0The coordinates of each final clustering central point in the table are respectively ordered with the ith clustering central point SiThe position offset epsilon between the coordinates of each cluster center point at the same position of the middle index valuejAnd an amount of angular offset
Figure GDA0003568086610000081
Wherein:
Figure GDA0003568086610000082
Figure GDA0003568086610000083
in the formula
Figure GDA0003568086610000084
For the 0 th cluster central point ordered table S0The abscissa and ordinate of the center point of each final cluster having a medium index value of j,
Figure GDA0003568086610000085
ordered list S for representing ith cluster center pointiThe abscissa and ordinate of each clustering center point with the middle index value of j, and F represents the amplification factor between the image length and the actual length, which can be flat according to the lens of the CCD cameraThe distance between the surface and the surface of the structure is selected, the larger the distance is, the larger the value of F is, and the general value range is 2-20.
Secondly, obtaining each cluster central point ordered list S by utilizing the math toolkit in the matlabiEach final cluster center point and 0 th cluster center point in the ordered list S0The maximum position offset epsilon of the structure surface is compared with the final clustering center point of each same index valuemaxAnd an amount of angular offset
Figure GDA0003568086610000086
Then, the data is sorted by the cluster central point SiThe shooting time corresponding to the gray-scale map of the laser speckle of the surface of the structure respectively corresponds to X-axis coordinates, and the maximum position offset epsilon of the surface of the structure corresponding to each cluster central point ordered tablemaxAnd an amount of angular offset
Figure GDA0003568086610000087
Plotting a line graph of the offset over time for the ordinate, wherein:
εmax＝MAX(ε1，...，εj...，εk)
Figure GDA0003568086610000091
in the formula epsilon1...εj...εkIndicating the position offset of all cluster center point coordinates at the ith time,
Figure GDA0003568086610000092
and the angular offset of the coordinates of the center points of all clusters at the ith moment is represented.
And step six, in the line graph of which the offset changes along with time in the step five, if the maximum position offset and the maximum angle offset of the surface of the structure are close to the preset critical deformation, the curve slope is overlarge, the curve slope is suddenly changed and the like, which indicates that the surface of the structure generates larger deformation in the lifting process, the lifting work is immediately stopped, and the related fixed lifting points (hanging rings welded on the surface of the structure, supporting points used for hydraulic lifting and the like) are fixed and reinforced by adopting the existing method, so that the lifting is continued after the lifting standard is reached.

Claims (1)

1. A method for monitoring the maximum deformation in the lifting process of a large-scale structure is characterized by comprising the following steps:
step one, collecting a gray level image of an initial state when the surface of a structure is not lifted and a state in the lifting process by using a CCD camera, wherein the specific process is as follows:
the method comprises the following steps that firstly, laser is adopted to irradiate the surface of a structure, so that a plurality of laser speckles which are approximately circular are generated on the surface of the structure, the central lines of the laser speckles are parallel to each other and are uniformly distributed on the surface of the structure, and the position of a CCD camera is adjusted, so that the lens plane of the CCD camera is parallel to the surface of the structure which generates the laser speckles;
when the structure is not lifted and in the lifting process, shooting the surface of the structure generating the laser speckles by using a CCD camera according to a set time interval to obtain a gray-scale image of the laser speckles on the surface of the structure at different moments;
and secondly, performing data extraction and data storage on each gray-scale image obtained by the CCD camera by using SQL Server software, wherein the specific process is as follows:
firstly, SQL Server software is used for creating an image information database, and then n laser speckle pixel point position tables A are established in the image information databaseiN cluster central point disorder table UiAnd n cluster central points ordered table SiI is 1 … n, and the tables with the same subscript i have corresponding relation with each other;
secondly, transmitting the gray level image obtained by the CCD camera to a computer externally connected with an image acquisition card through a network or a circuit;
thirdly, the computer reads the gray level images of the laser speckles on the surface of the structure at different moments by using an image acquisition card, establishes an image coordinate system with the coordinate origin positioned at the lower left corner of the image in each gray level image, records the total number k of the laser speckles, selects a pixel point in each laser speckle image according to a set sequence as an approximate substitution position of the laser speckle position and extracts selected pixel point coordinate data, and sequentially and respectively stores each pixel point coordinate data selected from each gray level image into a laser speckle pixel point position table corresponding to each gray level image in a list form;
step three, utilizing a K-means module in Python to respectively perform a table A of the positions of the laser speckle pixel pointsiClustering the coordinate data of each pixel point, wherein the specific process is as follows:
firstly, leading in a K-means module in Python, establishing connection with an SQL Server database, and extracting a laser speckle pixel point position table Ai；
Secondly, the position table A of each laser speckle pixel point is calculatediDividing all pixel point coordinate data into K feature spaces with similar sizes in sequence from front to back in a list by using a K-means module respectively, and randomly selecting a point in each feature space as an initial clustering central point;
thirdly, respectively calculating a position table A of each laser speckle pixel pointiThe distance from all pixel points to each initial clustering center point is compared, then the distance from each pixel point to each initial clustering center point is compared, each pixel point closest to each initial clustering center point is recombined with the initial clustering center point to form a reconstruction feature space, k reconstruction feature spaces are formed,
fourthly, calculating a position table A of each laser speckle pixel pointiMean value of horizontal and vertical coordinates of all pixel points in each reconstruction characteristic space
Figure FDA0003568086600000021
And
Figure FDA0003568086600000022
formed point
Figure FDA0003568086600000023
As the coordinates of the new cluster center point of each reconstructed feature space;
fifthly, adopting a position table A of each laser speckle pixel pointiThe coordinates of the new clustering center point of each reconstructed feature space replace the coordinates of the initial clustering center point in the third step, and the third step and the fifth step are repeated until the laser speckle pixel point position table AiThe coordinates of the new clustering center point in each reconstructed feature space are not changed any more, and the coordinates of the final clustering center point are obtained;
sixthly, the position table A of each laser speckle pixel point in the fifth stepiThe coordinates of each final clustering central point in the cluster central point list are respectively stored into a clustering central point disorder table U in an SQL database corresponding to the laser speckle pixel point position tableiPerforming the following steps;
step four, using SQL Server software to carry out data sequencing and index correspondence to the cluster center point unordered table, and the concrete process is as follows:
firstly, SQL Server software is adopted to arrange a table S in each cluster central pointiAllocating k blank storage areas in the middle, and sequentially allocating 1 … k index values;
step two, respectively calculating the ith cluster central point disorder table UiIn each final clustering central point and the (i-1) th clustering central point ordered table Si-1The distance between the k coordinate points is sorted with the i-1 st cluster center pointi-1The ith cluster center point disorder table U with each coordinate point nearest to the coordinate pointiThe final clustering center point coordinate in the table S is sorted in the ith-1 clustering center point according to the final clustering center pointi-1The index value in the table is stored in the same index value position in the ith cluster central point ordered table; wherein, i is 0 and represents the structure is not lifted, the 0 th cluster central point unordered table U0The coordinates of the final cluster central point in the table are directly and sequentially stored into the 0 th cluster central point ordered table S0Performing the following steps;
step five, drawing the image of the offset of the deformation area on the surface of the structure by utilizing matlab software, wherein the specific process is as follows:
first, using 0 th clustering central point to order table S0Taking the coordinates of each final clustering center point as a reference, and calculating the 0 th clustering center point ordered list S by using matlab software0The coordinates of each final clustering center point in the table are respectively compared with the ith clustering center point ordered table SiThe position offset epsilon between the coordinates of each cluster center point at the same position of the middle index valuejAnd an amount of angular offset
Figure FDA0003568086600000031
Secondly, obtaining each cluster central point ordered list S by utilizing the math toolkit in the matlabiEach final cluster central point and 0 th cluster central point ordered table S0The maximum position offset epsilon of the structure surface is compared with the final clustering center point of each same index valuemaxAnd an amount of angular offset
Figure FDA0003568086600000032
Then, the data is sorted by the cluster central point SiThe shooting time corresponding to the gray-scale map of the laser speckle of the surface of the structure respectively corresponds to X-axis coordinates, and the maximum position offset epsilon of the surface of the structure corresponding to each cluster central point ordered tablemaxAnd an amount of angular offset
Figure FDA0003568086600000033
Drawing a line graph of the change of the offset with time for the ordinate;
and step six, in the line graph of the deviation amount changing along with the time in the step five, if the maximum position deviation amount and the maximum angle deviation amount of the surface of the structure are close to the preset critical deformation amount, the curve slope is overlarge and the curve slope generates a sudden change, the lifting work is immediately stopped, the related fixed lifting point is fixedly strengthened, and the lifting is continued after the lifting standard is reached.
CN202011316723.4A 2020-11-23 2020-11-23 Method for monitoring maximum deformation in lifting process of large-scale structure Active CN112329705B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011316723.4A CN112329705B (en) 2020-11-23 2020-11-23 Method for monitoring maximum deformation in lifting process of large-scale structure

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011316723.4A CN112329705B (en) 2020-11-23 2020-11-23 Method for monitoring maximum deformation in lifting process of large-scale structure

Publications (2)

Publication Number Publication Date
CN112329705A CN112329705A (en) 2021-02-05
CN112329705B true CN112329705B (en) 2022-06-24

Family

ID=74321964

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011316723.4A Active CN112329705B (en) 2020-11-23 2020-11-23 Method for monitoring maximum deformation in lifting process of large-scale structure

Country Status (1)

Country Link
CN (1) CN112329705B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113532270B (en) * 2021-05-31 2022-07-08 天津博迈科海洋工程有限公司 Attitude detection method for FPSO flare tower installation process
CN113970552B (en) * 2021-09-26 2024-06-14 北京京仪仪器仪表研究总院有限公司 Apple nondestructive testing method combining laser speckle and Kmeans clustering algorithm

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102506738A (en) * 2011-11-08 2012-06-20 中国矿业大学 Real-time monitoring and early-warning method for deformation of geotechnical engineering based on digital photogrammetry
CN108759699A (en) * 2018-03-27 2018-11-06 西安交通大学 A kind of measurement method and system of big visual field masonry structure material three-dimensional whole field deformation
CN108918271A (en) * 2018-09-11 2018-11-30 苏州大学 Young's modulus measurement method based on microoptic digital speckle method
CN109883333A (en) * 2019-03-14 2019-06-14 武汉理工大学 A kind of non-contact displacement strain measurement method based on characteristics of image identification technology
CN110823116A (en) * 2019-10-25 2020-02-21 同济大学 Image-based building component deformation measurement method

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103575227B (en) * 2013-09-23 2016-07-06 西安新拓三维光测科技有限公司 A kind of vision extensometer implementation method based on digital speckle

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102506738A (en) * 2011-11-08 2012-06-20 中国矿业大学 Real-time monitoring and early-warning method for deformation of geotechnical engineering based on digital photogrammetry
CN108759699A (en) * 2018-03-27 2018-11-06 西安交通大学 A kind of measurement method and system of big visual field masonry structure material three-dimensional whole field deformation
CN108918271A (en) * 2018-09-11 2018-11-30 苏州大学 Young's modulus measurement method based on microoptic digital speckle method
CN109883333A (en) * 2019-03-14 2019-06-14 武汉理工大学 A kind of non-contact displacement strain measurement method based on characteristics of image identification technology
CN110823116A (en) * 2019-10-25 2020-02-21 同济大学 Image-based building component deformation measurement method

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
Dynamic Deformation Measurement of Specular Surface with Deflectometry and Speckle Digital Image Correlation;Yao Hu，et al;《Sensors》;20200226;全文 *
基于数字激光散斑图像的位移测量的多层平截算法;邱天，等;《光电子·激光》;20060930;第17卷(第6期);全文 *
散斑位移测量中一种新的计算窗口选取方法;曹慧;《工具技术》;20101231;第44卷(第12期);全文 *

Also Published As

Publication number Publication date
CN112329705A (en) 2021-02-05

Similar Documents

Publication Publication Date Title
CN112329705B (en) Method for monitoring maximum deformation in lifting process of large-scale structure
CN102003938B (en) Thermal state on-site detection method for large high-temperature forging
CN110675376A (en) PCB defect detection method based on template matching
CN106017350A (en) Machine-vision-based rapid detection device and detection method for medium and small module gears
CN104408055B (en) Storage method and device for laser radar point cloud data
Liu et al. An improved online dimensional measurement method of large hot cylindrical forging
CN109765240B (en) Device and method for detecting stitch defects of industrial parts
US20100021068A1 (en) System and method for extracting boundary elements of an object
CN109751964B (en) High-precision non-contact pipe diameter measuring method and device
CN110763136B (en) High-precision three-dimensional workpiece size detection system and method
CN109657607A (en) A kind of human face target distance measuring method, device and storage medium based on recognition of face
CN102607446A (en) Method for measuring steel structure welding deformation based on gridding
CN106124034A (en) Thin-wall part operation mode based on machine vision test device and method of testing
CN113670931A (en) Steel plate surface defect detection method and system based on neural network
CN116642897A (en) Aviation spare part coating appearance defect detection system
CN113095447B (en) Detection method and system based on image recognition
CN114199160A (en) Circuit board component geometric detection method based on binary coding grating defocused projection
CN112990190A (en) Method and device for identifying reading of pointer instrument
CN115389516A (en) Pavement disease de-weighting method based on high-precision positioning and deep neural network
CN113983951B (en) Three-dimensional target measuring method, device, imager and storage medium
CN114210741B (en) Strip steel surface roughness online detection device and method
CN115482206A (en) Method for detecting overall size of ultra-large type reinforcing steel bar net piece
CN111550273B (en) Hydraulic support leveling and straightening method based on computer vision
Samusev et al. Shaping of pipe blanks in the edge-bending press of the TESA 1420 pipe-electrowelding line
CN102708588A (en) Three-dimensional reconstruction method for damaged corrosive pitting topography of metal panel

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