CN110737652A - Data cleaning method and system for three-dimensional digital model of surface mine and storage medium - Google Patents

Data cleaning method and system for three-dimensional digital model of surface mine and storage medium Download PDF

Info

Publication number
CN110737652A
CN110737652A CN201910942073.5A CN201910942073A CN110737652A CN 110737652 A CN110737652 A CN 110737652A CN 201910942073 A CN201910942073 A CN 201910942073A CN 110737652 A CN110737652 A CN 110737652A
Authority
CN
China
Prior art keywords
matrix
sampling
foreground
equipment
elevation
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
Application number
CN201910942073.5A
Other languages
Chinese (zh)
Other versions
CN110737652B (en
Inventor
贾明滔
王佳恒
王李管
毕林
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.)
Central South University
Original Assignee
Central South University
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 Central South University filed Critical Central South University
Priority to CN201910942073.5A priority Critical patent/CN110737652B/en
Publication of CN110737652A publication Critical patent/CN110737652A/en
Application granted granted Critical
Publication of CN110737652B publication Critical patent/CN110737652B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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
    • G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G06T17/05—Geographic models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Geometry (AREA)
  • Software Systems (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Graphics (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Remote Sensing (AREA)
  • General Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Complex Calculations (AREA)

Abstract

The invention discloses a data cleaning method, a system and a storage medium for surface mine three-dimensional digital models, wherein the method comprises the steps of obtaining surface elevation values of a surface mine area, and then establishing an elevation matrix of the surface mine area, wherein the surface mine area comprises equipment and a pit area where the equipment is located, each matrix element in the elevation matrix corresponds to a position coordinate and an elevation value of each position in the surface mine area, then carrying out standard deviation sampling on the elevation matrix to obtain a standard deviation sampling matrix, and adopting a top cap algorithm to convert the standard deviation sampling matrix to obtain a foreground-background separation matrix, wherein the equipment is used as a foreground, the pit area where the equipment is located is used as a background, then obtaining the matrix elements corresponding to the equipment by using the foreground-background separation matrix, obtaining the position coordinates and the elevation values of the equipment according to the matrix elements, and finally carrying out interpolation replacement on the equipment elevation values according to an equipment position sampling interpolation method.

Description

Data cleaning method and system for three-dimensional digital model of surface mine and storage medium
Technical Field
The invention belongs to the field of open-pit mining, and particularly relates to a data cleaning method, a data cleaning system and a storage medium for open-pit mine three-dimensional digital models.
Background
In the field of surface mining, the computer technology is utilized to realize the digitization, automation and informatization of mines, and the digitization technology is applied to the production process of surface mining, which is beneficial to improving the production efficiency and the safety level.
However, although the three-dimensional digital model obtained by the currently common digital surface model obtaining method (global positioning system real-time dynamic measurement, unmanned airborne laser radar measurement, unmanned aerial vehicle oblique photogrammetry, etc.) can highly restore the original appearance of the mine, the data volume of the model is large and the model contains a large amount of interference information, such as the spatial information of mining equipment, the spatial information of mine structures, etc. due to the interference, when the three-dimensional digital model of the open pit is used for subsequent planning, mining design, etc., the open pit terrain control line is often required to be manually depicted and extracted, so that the model is optimized.
Disclosure of Invention
The invention aims to provide data cleaning methods, systems and storage media for a three-dimensional digital model of a surface mine, which can automatically identify the position of mining equipment in the three-dimensional digital model of the surface mine and filter the mining equipment.
The invention provides a data cleaning method of surface mine three-dimensional digital models, which comprises the following steps:
step S1: acquiring a surface elevation value of a surface mine area based on a three-dimensional digital model of the surface mine, and then establishing an elevation matrix of the surface mine area;
the surface mine area comprises equipment and a pit area where the equipment is located, and each matrix element in the elevation matrix corresponds to a position coordinate and an elevation value of each position in the surface mine mountain area;
step S2: performing standard deviation sampling on the elevation matrix to obtain a standard deviation sampling matrix;
step S3: converting the standard deviation sampling matrix by adopting a top hat algorithm to obtain a foreground and background separation matrix, wherein the equipment is regarded as a foreground, and a pit area where the equipment is located is regarded as a background;
step S4: acquiring matrix elements corresponding to the equipment by using the foreground and background separation matrix, and acquiring position coordinates and elevation values of the equipment according to the matrix elements;
step S5: and carrying out interpolation replacement on the elevation value of the equipment according to an equipment position sampling interpolation method.
The method is based on the condition that interference of equipment data exists in the mine field, and the method realizes cleaning of the equipment data in the three-dimensional digital model, so that the three-dimensional digital model can more accurately restore the original appearance of the mine, wherein in the aspect, the method realizes data cleaning based on the three-dimensional model of the elevation matrix, compared with a conventional TIN model (triangulation network model), the data editing performance is stronger, and the automation of the data cleaning is easier to realize, although the efficiency of the triangulation network model on data storage is higher, the editing difficulty is large, and the automation degree is low, in the two aspects, the method discovers that the terrain change of the pit area of the open mine is smooth and regular, the dispersion degree of the elevation data is small, the edge elevation mutation of the equipment area is large, and the dispersion degree of the elevation data is large, so that the method skillfully utilizes a standard difference sampling matrix to primarily separate the equipment and the pit area based on the special terrain of the open mine, the standard difference data obtained by the equipment area is larger than the standard difference value of the pit area, thereby realizing the separation of the top pit and the equipment from the top slope surface of the initial separation by utilizing .
In summary, the values of the device and the pit area in the standard deviation sampling matrix and the foreground-background separation matrix are greatly different, so that the device and the pit area can be effectively distinguished, and the matrix elements corresponding to the device are extracted.
, preferably, the step S2 of sampling the elevation matrix with the standard deviation to obtain the standard deviation sampling matrix comprises the steps of sliding the sampling kernel C in the elevation matrix and obtaining the element values in the standard deviation sampling matrix;
wherein, element values in the standard deviation sampling matrix are calculated according to the standard deviation formula every times, and the standard deviation formula is as follows:
in the formula, SNRepresents element values, C, in the standard deviation sampling matrix corresponding to the current slipsijRepresenting the values of the ith row and jth column of the current sliding sampling core C,
Figure BDA0002223194220000022
and representing the average value of the elements in the sampling kernel C in current sliding, wherein the size of the sampling kernel C is mxn, and the element value in the sampling kernel C in the sliding process is the element value in the area where the sampling kernel C is located in the elevation matrix.
The sliding step of the sampling core C is, for example, m × n, but the value is not specifically limited in the present invention.
, preferably, in step S3, the standard deviation sampling matrix is converted to obtain a foreground and background separation matrix according to the following formula:
Figure BDA0002223194220000024
wherein, the structural elements of B (i, j) are as follows:
Figure BDA0002223194220000023
wherein T (X, Y) represents a foreground-background separation matrix, S (X, Y) represents a standard deviation sampling matrix, theta represents corrosion operation,
Figure BDA0002223194220000031
indicating the expansion operation, X, Y indicating the position coordinates, X the horizontal axis, Y the vertical axis.
The top hat algorithm may increase the contrast of the device and pit area, allowing the degree of dispersion to be further expanded, allowing the device to be effectively distinguished from the slope.
, preferably, the process of identifying the matrix element corresponding to the device by using the foreground and background separation matrix in the step S4 is to perform binarization processing based on the foreground and background separation matrix to obtain a binarization matrix, and then identify the matrix element corresponding to the device;
wherein, a threshold value of binarization processing is determined by adopting a maximum inter-class variance method;
and the elements larger than the threshold are matrix elements corresponding to the equipment, and the elements smaller than the threshold are matrix elements corresponding to the pit area where the equipment is located.
, preferably, before the binarization processing is carried out by using the foreground and background separation matrix to obtain the binarization matrix, regularizing the foreground and background separation matrix to obtain a regularized matrix;
wherein, the regularization treatment is carried out according to the following formula:
Figure BDA0002223194220000032
in the formula (I), the compound is shown in the specification,
Figure BDA0002223194220000033
representing a regularization matrix TNThe value of the element in the ith row and jth column in (X, Y), TijThe element values T of the ith row and the jth column in the foreground and background separation matrixmin、TmaxRespectively the minimum value and the maximum value in the foreground-background separation matrix.
, preferably, the process of determining the threshold value of the binarization processing by the maximum inter-class variance method is as follows:
firstly, values are sequentially taken as threshold values t in the element value range of a foreground and background separation matrix or a regularization matrix, the matrix is divided into a foreground class interval and a background class interval according to the element values, and then the variance sigma corresponding to each threshold value t is calculated2(t), wherein the element value range of the foreground class interval is [0, t ]]The element value range of the background class interval is [ t +1, L-1 ]]L-1 is the maximum value of the element values of the foreground and background separation matrix or the regularization matrix;
then, the variance σ is selected2(t) the maximum threshold value t is used as the threshold value of the binarization processing;
the variance σ2The calculation formula of (t) is as follows:
σ2(t)＝ω0(μ0-μ)2+ω1(μ1-μ)2
in the formula, ω0、ω1The occurrence probabilities of the foreground class and the background class are respectively equal to the ratio of the number of elements in the foreground class interval, the number of elements in the background class interval and the total number of elements in the foreground and background separation matrix or the regularization matrix;
μ0、μ1and mu is the mean value of the element values in the foreground class interval, the mean value of the element values in the background class interval, and the mean value of the element values in the foreground-background separation matrix or the regularization matrix respectively.
It should be understood that the larger the corresponding variance, the better the separation.
And , preferably, after the binarization processing is performed on the basis of the foreground and background separation matrix to obtain a binarization matrix, the method further comprises the steps of upsampling the binarization matrix to the size of the elevation matrix, and then identifying matrix elements corresponding to the equipment.
, preferably, in step S5, performing interpolation replacement on the elevation value of the device by using an inverse distance weighted interpolation method, where a sampling interpolation window is set in the device, a central point of the sampling interpolation window corresponds to the device position, and a sampling point other than the central point is a pit area point adjacent to the device;
the calculation formula of the inverse distance weighted interpolation is as follows:
Figure BDA0002223194220000041
wherein Z is an elevation value after interpolation replacement when the equipment is taken as a point to be interpolated, and Z isiIs the elevation of the sampling point, diThe distance between the point to be interpolated and the sampling point,and n is the number of sampling points in the sampling interpolation window.
In another aspect, the present invention provides systems based on the above methods, including:
an elevation matrix construction module: the method comprises the steps of obtaining a surface elevation value of a surface mine area based on a surface mine three-dimensional digital model, and then establishing an elevation matrix of the surface mine area;
a standard deviation sampling matrix construction module: the standard deviation sampling device is used for sampling the standard deviation of the elevation matrix to obtain a standard deviation sampling matrix;
the foreground and background separation matrix construction module: the foreground and background separation matrix is obtained by converting the standard deviation sampling matrix by adopting a top hat algorithm;
the equipment extraction module is used for acquiring matrix elements corresponding to the equipment by using the foreground and background separation matrix and acquiring the position coordinates and the elevation values of the equipment according to the matrix elements;
and the interpolation module is used for carrying out interpolation replacement on the elevation value of the equipment according to the equipment position sampling interpolation method.
Furthermore, the present invention also provides storage media storing computer program instructions that, when executed by a terminal device, cause the terminal device to perform the method of the claims above.
Advantageous effects
The method provided by the invention realizes automatic cleaning of the equipment data in the three-dimensional digital model based on the condition that the equipment data is interfered in the mine field, so that the three-dimensional digital model can more accurately restore the original appearance of the mine.
According to the method, the local elevation change of the mine pit is considered firstly, the position of the equipment is preliminarily determined through local standard deviation sampling, the initial separation of the equipment and the mine pit area can be realized by utilizing the standard deviation based on the fact that the terrain change of the mine pit area of the open-pit mine is smooth and regular, the dispersion degree of elevation data is small, the edge elevation of the equipment area is suddenly changed, and the dispersion degree of the elevation data is large, and then the equipment and the mine pit area are separated in steps by utilizing a top cap algorithm in mathematical morphology, particularly the slope terrain and the equipment in the mine pit area can be effectively distinguished, so that the reliability of final equipment position identification is improved, and the reliability of data cleaning is improved.
Drawings
FIG. 1 is a schematic flow chart of a data cleaning method for surface mine three-dimensional digital models, provided by an embodiment of the invention;
fig. 2 is a schematic diagram of a three-dimensional digital model and a corresponding ortho-image of a surface mine according to an embodiment of the present invention, wherein (a) is an ortho-image of the surface mine, and (b) is a three-dimensional digital surface model of the surface mine, and a white frame is selected where mining equipment is located;
FIG. 3 is a schematic diagram of a local standard deviation sample S (X, Y) according to an embodiment of the present invention;
fig. 4 is a schematic diagram of the foreground and background separation matrix T (X, Y) provided in the embodiment of the present invention;
fig. 5 is a schematic diagram of a binarization matrix B (X, Y) according to an embodiment of the present invention;
fig. 6 is a schematic diagram of an upsampling method provided by an embodiment of the present invention;
fig. 7 is a schematic diagram illustrating an effect of performing interpolation replacement on the elevation of the device based on an inverse distance weighted interpolation method according to the embodiment of the present invention.
Detailed Description
The present invention will now be described in further with reference to examples.
The surface mine three-dimensional digital model data cleaning method provided by the invention aims to extract and filter the main interference information in the three-dimensional digital model, namely the spatial information of mining equipment, to obtain the three-dimensional digital model only containing the ground elevation information of a mine pit area, and is specifically used for identifying, positioning and filtering the mining equipment in the surface mine three-dimensional digital model, namely effectively filtering the elevation value data of the equipment, eliminating the interference of the equipment and more highly restoring the original appearance of the mine.
As shown in fig. 1, the method for cleaning data of a three-dimensional digital model of a surface mine according to an embodiment of the present invention includes the following steps:
step 101: the surface elevation value of the surface mine area is obtained based on a surface mine three-dimensional digital model (DSM). Wherein the surface mine area includes equipment and a pit area in which the equipment is located. The specific process is as follows: the method comprises the steps of establishing a surface mine DSM (digital surface model) based on measurement technologies such as oblique photogrammetry technology or unmanned aerial vehicle-mounted LiDAR measurement technology; then acquiring three-dimensional point cloud information of the equipment and a pit area where the equipment is located based on a three-dimensional digital model DSM of the surface mine; and determining the surface elevation value of the surface mine area based on the equipment and the three-dimensional point cloud information of the pit area where the equipment is located. As shown in fig. 2, (a) is an orthographic view of a surface mine, and (b) is a surface mine DSM in which the white boxes are the locations of mining equipment.
Step 102: and constructing equipment and an elevation matrix Z (X, Y) of the pit area where the equipment is located based on the surface elevation values of the surface mine area. And each matrix element in the elevation matrix corresponds to the position coordinate and the elevation value of the equipment and each position in the pit area where the equipment is located. Z (X, Y) represents an elevation Z with position coordinates (X, Y), and XY is a Cartesian coordinate system, i.e., X is the horizontal axis and Y is the vertical axis.
Step 103: the elevation matrix Z (X, Y) is subjected to standard deviation sampling by using a local standard deviation sampling algorithm to obtain a standard deviation sampling matrix S (X, Y), as shown in fig. 3.
Calculating element values in the standard deviation sampling matrix according to a standard deviation formula every times, wherein the standard deviation formula is as follows:
in the formula, SNRepresents element values, C, in the standard deviation sampling matrix corresponding to the current slipsijRepresenting the values of the ith row and jth column of the current sliding sampling core C,
Figure BDA0002223194220000062
the average value of the elements in the sampling kernel C in current sliding is represented, the size of the sampling kernel C is mxn, and the element value in the sampling kernel C in the sliding process is the element value in the area where the sampling kernel C is located in the elevation matrix.
From the above, the size of the standard deviation matrix S (X, Y) is smaller than the size of the elevation matrix Z (X, Y) of the equipment and the pit area in which the equipment is located.
Step 104: and converting the standard deviation sampling matrix by adopting a top hat algorithm to obtain a foreground and background separation matrix T (X, Y), wherein the equipment is taken as a foreground, and a pit area where the equipment is located is taken as a background. Fig. 4 shows the foreground-background separation matrix T (X, Y).
The top cap algorithm corresponds to the following formula:
Figure BDA0002223194220000063
wherein, the structural elements of B (i, j) are as follows:
Figure BDA0002223194220000064
wherein T (X, Y) represents a foreground-background separation matrix, S (X, Y) represents a standard deviation sampling matrix, theta represents corrosion operation,
Figure BDA0002223194220000065
indicating the dilation operation.
Step 105: regularizing the foreground-background separation matrix T (X, Y) to obtain a regularization matrix TN(X, Y) and applying the maximum inter-class variance method to the regularization matrix TNAnd (X, Y) performing binarization conversion to obtain a binarization matrix B (X, Y) of the foreground and background separation matrix. As shown in fig. 5.
Wherein, the regularization treatment is carried out according to the following formula:
Figure BDA0002223194220000066
in the formula (I), the compound is shown in the specification,
Figure BDA0002223194220000067
representing a regularization matrix TNThe value of the element in the ith row and jth column in (X, Y), TijThe element values T of the ith row and the jth column in the foreground and background separation matrixmin、TmaxRespectively the minimum value and the maximum value in the foreground-background separation matrix.
As can be seen from the above regularization formula, the values of the elements in the regularization matrix are all within the interval of 0 to 255.
And determining a threshold value of binarization processing by adopting a maximum inter-class variance method, wherein the implementation process comprises the following steps:
first, regularization matrix TNSequentially taking values in the element value range of (X, Y) as threshold values t, dividing the matrix into foreground intervals and background intervals according to the element values, and calculating the variance sigma corresponding to each threshold value t2(t), wherein the element value range of the foreground class interval is [0, t ]]The element value range of the background class interval is [ t +1, L-1 ]]L-1 is the maximum value of the element values of the foreground and background separation matrix or the regularization matrix;
then, the variance σ is selected2(t) the maximum threshold value t is used as the threshold value for the binarization process.
The formula is as follows:
Th＝argmax{σ2(t)}；σ2(t)＝ω0(μ0-μ)2+ω1(μ1-μ)2
where Th is the threshold value of the binarization process, ω0、ω1The occurrence probabilities of the foreground class and the background class are respectively equal to the ratio of the number of elements in the foreground class interval to the total number of elements in the regularization matrix; mu.s0、μ1And mu are the mean value of the element values in the foreground class interval, the mean value of the element values in the background class interval and the mean value of the element values in the regularization matrix respectively.
The binarization process in this embodiment is as follows: regularizing matrix TNIn (X, Y), the label whose element value is greater than the threshold Th is 1, and the label whose element value is less than the threshold Th is 0, to obtain the binary matrix B (X, Y). In this embodiment, the binarization processing is performed based on the regularization matrix, and in other feasible embodiments, the binarization processing may be performed based on the foreground-background separation matrix.
Step 106: and (3) upsampling the binarization matrix B (X, Y) to the size of the elevation matrix Z (X, Y), and extracting matrix elements with elements of 1 in the matrix B (X, Y), thereby acquiring the coordinates (X, Y) of the equipment and the elevation value Z.
As shown in fig. 6, in this embodiment, the upsampling method is as follows: taking an example of upsampling a matrix with the size of 2 × 2 to a matrix with the size of 6 × 6, taking each element in the original 2 × 2 matrix as a center, constructing a 6 × 6 sparse matrix, filling other element values in the sparse matrix with each center point value, and obtaining the matrix, namely the 6 × 6 upsampling matrix.
Step 107: and acquiring elevation values of the equipment and the pit areas adjacent to the equipment according to the equipment coordinates, and performing interpolation replacement on the elevation values of the equipment by taking the elevation values of the pit areas adjacent to the equipment as a reference based on an inverse distance weighting interpolation method.
The method comprises the steps that a sampling interpolation window is arranged at the position of equipment, the central point of the sampling interpolation window corresponds to the position of the equipment, and the sampling point of a non-central point is a pit area point adjacent to the equipment. The method specifically comprises the following steps: acquiring an edge elevation value of the equipment, and taking the edge elevation value as a point to be interpolated; constructing a sampling interpolation window by taking the elevation value of the edge of the equipment as a center, and searching the neighborhood of the sampling interpolation window; and performing inverse distance weighted interpolation calculation on the obtained sampling points to replace the elevation value of the equipment. In this embodiment, the sampling interpolation window size is 3 × 3.
The calculation formula of the inverse distance weighted interpolation is as follows:
Figure BDA0002223194220000071
wherein Z is an elevation value after interpolation replacement when the equipment is taken as a point to be interpolated, and Z isiIs the elevation of the sampling point, diThe distance between the point to be interpolated and the sampling point,
Figure BDA0002223194220000081
and n is the number of sampling points in the sampling interpolation window. In this embodiment, the distance d can be simplified to be the distance from each point in the 3 x 3 square window to the center point, i.e. 1 or
Figure BDA0002223194220000082
The weight u typically takes 2.
It should be understood that the invention identifies the position of the equipment and replaces the elevation value of the equipment, so that the data in the three-dimensional digital model of the surface mine is the data of the pit area, and the pit can be restored more truly.
According to the embodiment of the invention, the local elevation change of a mine pit is firstly inspected, the position of the equipment is preliminarily determined through local standard deviation sampling, then the equipment and the mine pit are separated by using a top cap algorithm in mathematical morphology, steps are carried out on the equipment through a maximum inter-class variance method, noise points are filtered, and the coordinates of the equipment are finally extracted.
Based on the method, the embodiment of the invention also provides a data cleaning system for surface mine three-dimensional digital models, which comprises the following steps:
an elevation matrix construction module: the method comprises the steps of obtaining a surface elevation value of a surface mine area based on a surface mine three-dimensional digital model, and then establishing an elevation matrix of the surface mine area;
a standard deviation sampling matrix construction module: the standard deviation sampling device is used for sampling the standard deviation of the elevation matrix to obtain a standard deviation sampling matrix;
the foreground and background separation matrix construction module: the foreground and background separation matrix is obtained by converting the standard deviation sampling matrix by adopting a top hat algorithm;
a regularization module: and the regularization matrix is obtained by regularizing the foreground and background separation matrix.
And the equipment extraction module is used for acquiring matrix elements corresponding to the equipment by using the foreground and background separation matrix and then acquiring the position coordinates and the elevation values of the equipment according to the matrix elements. In this embodiment, a maximum inter-class variance method is specifically adopted to determine a threshold value of binarization processing, construct a binarization matrix, and then obtain a device position coordinate and an elevation value.
And the interpolation module is used for carrying out interpolation replacement on the elevation value of the equipment according to the equipment position sampling interpolation method.
It should be understood that the functional unit modules in the embodiments of the present invention may be centralized in processing units, or each unit module may exist alone physically, or two or more unit modules are integrated in unit modules, and may be implemented in a form of hardware or software.
The present invention also provides storage media storing computer program instructions that, when executed by a terminal device, cause the terminal device to execute the method for cleaning surface mine three-dimensional digital model data described above.
Moreover, the present application may take the form of a computer program product embodied on or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
It is to be understood that each flow and/or block in the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions which can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagram flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

Claims (10)

  1. The data cleaning method of the three-dimensional digital model of the surface mines is characterized by comprising the following steps:
    step S1: acquiring a surface elevation value of a surface mine area based on a three-dimensional digital model of the surface mine, and then establishing an elevation matrix of the surface mine area;
    the surface mine area comprises equipment and a pit area where the equipment is located, and each matrix element in the elevation matrix corresponds to a position coordinate and an elevation value of each position in the surface mine mountain area;
    step S2: performing standard deviation sampling on the elevation matrix to obtain a standard deviation sampling matrix;
    step S3: converting the standard deviation sampling matrix by adopting a top hat algorithm to obtain a foreground and background separation matrix, wherein the equipment is regarded as a foreground, and a pit area where the equipment is located is regarded as a background;
    step S4: acquiring matrix elements corresponding to the equipment by using the foreground and background separation matrix, and acquiring position coordinates and elevation values of the equipment according to the matrix elements;
    step S5: and carrying out interpolation replacement on the elevation value of the equipment according to an equipment position sampling interpolation method.
  2. 2. The method of claim 1, wherein: the process of sampling the elevation matrix by the standard deviation to obtain the standard deviation sampling matrix in step S2 is as follows: sliding a sampling core C in the elevation matrix and acquiring element values in a standard deviation sampling matrix;
    wherein, element values in the standard deviation sampling matrix are calculated according to the standard deviation formula every times, and the standard deviation formula is as follows:
    Figure FDA0002223194210000011
    in the formula, SNRepresents element values, C, in the standard deviation sampling matrix corresponding to the current slipsijRepresenting the values of the ith row and jth column of the current sliding sampling core C,
    Figure FDA0002223194210000013
    and representing the average value of the elements in the sampling kernel C in current sliding, wherein the size of the sampling kernel C is mxn, and the element value in the sampling kernel C in the sliding process is the element value in the area where the sampling kernel C is located in the elevation matrix.
  3. 3. The method of claim 1, wherein: in step S3, the standard deviation sampling matrix is converted to obtain a foreground-background separation matrix according to the following formula:
    Figure FDA0002223194210000015
    wherein, the structural elements of B (i, j) are as follows:
    Figure FDA0002223194210000012
    wherein T (X, Y) represents a foreground-background separation matrix, S (X, Y) represents a standard deviation sampling matrix, theta represents corrosion operation,
    Figure FDA0002223194210000014
    indicating the expansion operation, X, Y indicating the position coordinates, X the horizontal axis, Y the vertical axis.
  4. 4. The method of claim 1, wherein: the process of identifying the matrix element corresponding to the device using the foreground-background separation matrix in step S4 is as follows: carrying out binarization processing based on the foreground and background separation matrix to obtain a binarization matrix, and then identifying matrix elements corresponding to equipment;
    wherein, a threshold value of binarization processing is determined by adopting a maximum inter-class variance method;
    the elements larger than or equal to the threshold are matrix elements corresponding to the equipment, and the elements smaller than the threshold are matrix elements corresponding to the pit area where the equipment is located.
  5. 5. The method of claim 4, wherein: before the binarization processing is carried out by using the foreground and background separation matrix to obtain the binarization matrix, the method also comprises the following steps: regularizing the foreground and background separation matrix to obtain a regularized matrix;
    wherein, the regularization treatment is carried out according to the following formula:
    Figure FDA0002223194210000021
    in the formula (I), the compound is shown in the specification,
    Figure FDA0002223194210000022
    representing a regularization matrix TNThe value of the element in the ith row and jth column in (X, Y), TijThe element values T of the ith row and the jth column in the foreground and background separation matrixmin、TmaxRespectively the minimum value and the maximum value in the foreground-background separation matrix.
  6. 6. The method according to claim 4 or 5, characterized in that: the process of determining the threshold value of the binarization processing by adopting the maximum inter-class variance method is as follows:
    firstly, values are sequentially taken as threshold values t in the element value range of a foreground and background separation matrix or a regularization matrix, the matrix is divided into a foreground class interval and a background class interval according to the element values, and then the variance sigma corresponding to each threshold value t is calculated2(t), wherein the element value range of the foreground class interval is [0, t ]]The element value range of the background class interval is [ t +1, L-1 ]]L-1 is the maximum value of the element values of the foreground and background separation matrix or the regularization matrix;
    then, the variance σ is selected2(t) the maximum threshold value t is used as the threshold value of the binarization processing;
    the variance σ2The calculation formula of (t) is as follows:
    σ2(t)＝ω0(μ0-μ)2+ω1(μ1-μ)2
    in the formula, ω0、ω1The occurrence probabilities of the foreground class and the background class are respectively equal to the ratio of the number of elements in the foreground class interval, the number of elements in the background class interval and the total number of elements in the foreground and background separation matrix or the regularization matrix;
    μ0、μ1and mu is the mean value of the element values in the foreground class interval, the mean value of the element values in the background class interval, and the mean value of the element values in the foreground-background separation matrix or the regularization matrix respectively.
  7. 7. The method of claim 4, wherein: after carrying out binarization processing on the basis of the foreground and background separation matrix to obtain a binarization matrix, the method further comprises the following steps: and (4) up-sampling the binary matrix to the size of the elevation matrix, and then identifying matrix elements corresponding to the equipment.
  8. 8. The method of claim 1, wherein: in the step S5, performing interpolation replacement on the elevation value of the device by using an inverse distance weighted interpolation method, wherein a sampling interpolation window is set in the device, the central point of the sampling interpolation window corresponds to the device position, and the sampling point of the non-central point is a pit area point adjacent to the device;
    the calculation formula of the inverse distance weighted interpolation is as follows:
    Figure FDA0002223194210000031
    wherein Z is an elevation value after interpolation replacement when the equipment is taken as a point to be interpolated, and Z isiIs the elevation of the sampling point, diThe distance between the point to be interpolated and the sampling point,
    Figure FDA0002223194210000032
    and n is the number of sampling points in the sampling interpolation window.
  9. 9, system based on the method claimed in any of claims 1-8, , comprising:
    an elevation matrix construction module: the method comprises the steps of obtaining a surface elevation value of a surface mine area based on a surface mine three-dimensional digital model, and then establishing an elevation matrix of the surface mine area;
    a standard deviation sampling matrix construction module: the standard deviation sampling device is used for sampling the standard deviation of the elevation matrix to obtain a standard deviation sampling matrix;
    the foreground and background separation matrix construction module: the foreground and background separation matrix is obtained by converting the standard deviation sampling matrix by adopting a top hat algorithm;
    the equipment extraction module is used for acquiring matrix elements corresponding to the equipment by using the foreground and background separation matrix and acquiring the position coordinates and the elevation values of the equipment according to the matrix elements;
    and the interpolation module is used for carrying out interpolation replacement on the elevation value of the equipment according to the equipment position sampling interpolation method.
  10. Storage medium 10, , storing computer program instructions which, when executed by a terminal device, cause the terminal device to perform the method of any of claims 1-8 to .
CN201910942073.5A 2019-09-30 2019-09-30 Data cleaning method and system for three-dimensional digital model of surface mine and storage medium Active CN110737652B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910942073.5A CN110737652B (en) 2019-09-30 2019-09-30 Data cleaning method and system for three-dimensional digital model of surface mine and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910942073.5A CN110737652B (en) 2019-09-30 2019-09-30 Data cleaning method and system for three-dimensional digital model of surface mine and storage medium

Publications (2)

Publication Number Publication Date
CN110737652A true CN110737652A (en) 2020-01-31
CN110737652B CN110737652B (en) 2022-03-11

Family

ID=69268386

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910942073.5A Active CN110737652B (en) 2019-09-30 2019-09-30 Data cleaning method and system for three-dimensional digital model of surface mine and storage medium

Country Status (1)

Country Link
CN (1) CN110737652B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101900546A (en) * 2009-05-27 2010-12-01 中国科学院地理科学与资源研究所 Construction method for digital elevation model for discrete expression of landform on earth surface
CN105893972A (en) * 2016-04-08 2016-08-24 深圳市智绘科技有限公司 Automatic illegal building monitoring method based on image and realization system thereof
CN106199557A (en) * 2016-06-24 2016-12-07 南京林业大学 A kind of airborne laser radar data vegetation extracting method
CN109816707A (en) * 2018-12-25 2019-05-28 中铁第四勘察设计院集团有限公司 A kind of field of opencast mining information extracting method based on high-resolution satellite image
US10360900B1 (en) * 2011-07-03 2019-07-23 Reality Analytics, Inc. System and method for taxonomically distinguishing sample data captured from sources

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101900546A (en) * 2009-05-27 2010-12-01 中国科学院地理科学与资源研究所 Construction method for digital elevation model for discrete expression of landform on earth surface
US10360900B1 (en) * 2011-07-03 2019-07-23 Reality Analytics, Inc. System and method for taxonomically distinguishing sample data captured from sources
CN105893972A (en) * 2016-04-08 2016-08-24 深圳市智绘科技有限公司 Automatic illegal building monitoring method based on image and realization system thereof
CN106199557A (en) * 2016-06-24 2016-12-07 南京林业大学 A kind of airborne laser radar data vegetation extracting method
CN109816707A (en) * 2018-12-25 2019-05-28 中铁第四勘察设计院集团有限公司 A kind of field of opencast mining information extracting method based on high-resolution satellite image

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
JINMIAO WANG 等: "Implicit 3D Modeling of Ore Body from Geological Boreholes Data Using Hermite Radial Basis Functions", 《MINERALS》 *
MOHAMMAD SAYAB 等: "Virtual Structural Analysis of Jokisivu Open Pit Using "Structure-from-Mition"Unmanned Aeral Vehicles (UAV) Photogrammetry:Implications for Structurally-Controlled Gold Deposits in Southwest Finland", 《REMOTE SENSING》 *
张炬 等: "露天矿境界优化几何约束模型优化及其应用", 《黄金科学技术》 *

Also Published As

Publication number Publication date
CN110737652B (en) 2022-03-11

Similar Documents

Publication Publication Date Title
CN111784685B (en) Power transmission line defect image identification method based on cloud edge cooperative detection
US20230084869A1 (en) System for simplified generation of systems for broad area geospatial object detection
US10970543B2 (en) Distributed and self-validating computer vision for dense object detection in digital images
CN108681692B (en) Method for identifying newly added buildings in remote sensing image based on deep learning
US9846975B2 (en) Generating filtered, three-dimensional digital ground models utilizing multi-stage filters
CN109272509B (en) Target detection method, device and equipment for continuous images and storage medium
CN110598541B (en) Method and equipment for extracting road edge information
CN113936256A (en) Image target detection method, device, equipment and storage medium
CN107564009B (en) Outdoor scene multi-target segmentation method based on deep convolutional neural network
CN106295613A (en) A kind of unmanned plane target localization method and system
US10685443B2 (en) Cloud detection using images
CN115797350B (en) Bridge disease detection method, device, computer equipment and storage medium
CN101710422B (en) Image segmentation method based on overall manifold prototype clustering algorithm and watershed algorithm
CN115240149A (en) Three-dimensional point cloud detection and identification method and device, electronic equipment and storage medium
CN110852327A (en) Image processing method, image processing device, electronic equipment and storage medium
JP2017111814A (en) Recognition method, apparatus and selection system by equipment for deposit
CN110717496A (en) Complex scene tree detection method based on neural network
CN114519819B (en) Remote sensing image target detection method based on global context awareness
CN113393385A (en) Unsupervised rain removal method, system, device and medium based on multi-scale fusion
CN112734675A (en) Image rain removing method based on pyramid model and non-local enhanced dense block
CN110737652B (en) Data cleaning method and system for three-dimensional digital model of surface mine and storage medium
CN116543333A (en) Target recognition method, training method, device, equipment and medium of power system
CN116310832A (en) Remote sensing image processing method, device, equipment, medium and product
CN113920273B (en) Image processing method, device, electronic equipment and storage medium
CN113963178A (en) Method, device, equipment and medium for detecting infrared dim and small target under ground-air background

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