CN108182440A - It is a kind of that the method for obtaining surrounding rock category is identified based on slag picture - Google Patents
It is a kind of that the method for obtaining surrounding rock category is identified based on slag picture Download PDFInfo
- Publication number
- CN108182440A CN108182440A CN201810019670.6A CN201810019670A CN108182440A CN 108182440 A CN108182440 A CN 108182440A CN 201810019670 A CN201810019670 A CN 201810019670A CN 108182440 A CN108182440 A CN 108182440A
- Authority
- CN
- China
- Prior art keywords
- slag
- surrounding rock
- grain size
- obtains
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Probability & Statistics with Applications (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of method for being identified based on slag picture and obtaining surrounding rock category, step is as follows：S1 obtains slag picture；S2, the slag picture that obtains that treated；S3, the Grades of Surrounding Rock association sensitive features collection of slag picture after calculation processing；Data sample is divided k cluster by S4 using AP clustering methods；S5 does LSSVM recurrence by gaussian kernel function and Polynomial kernel function Weighted Fusion, and to each cluster in step S4, obtains k submodel；K obtained submodel is weighted fusion, obtains surrounding rock category value by S6.The present invention is acquired realtime graphic processing, without manually participating in, avoids, excessively by artificial experience, improving the accuracy and rapidity of fender graded.
Description
Technical field
The invention belongs to tunnel surrounding technical field of image processing, and in particular to one kind is obtained based on the identification of slag picture encloses
Rock class method for distinguishing.
Background technology
Tunnel surrounding grade separation has vital effect in constructing tunnel, is the judgement of tunnel excavation rear stability
And corresponding supporting measure an important factor for choosing.Intensity of traditional fender graded according to rock, the integrality of rock mass, crack is filled out
Depending on situations such as filling object, underground water and crustal stress synthesis, since workload is larger and is difficult to obtain, lead to enclosing in reconnoitring early period
Rock classification is more rough.Therefore, in TBM tunneling processes, often occur disclosing rock mass and aforementioned obtained rock mass country rock point
The inconsistent situation of grade.In addition, due to the presence of TBM cutterheads and shield, acquisition face geological condition is limited after excavation, in real time
It is more difficult to obtain country rock grade.In site operation, rock mass fender graded is not according to the side in national standard specification in driving
Method defines, it is more by have exploration report with depending on field engineer's experience, it is understood that there may be divide wrong with lag
Problem.This has larger restriction for the accurate judgement of front rock mass conditions.Therefore, it finds a kind of real-time and accurately judges rock
The method of body fender graded is extremely important.
Invention content
Deficiency described in for the above-mentioned prior art, the present invention are provided a kind of identified based on slag picture and obtain country rock class
Method for distinguishing, fender graded lag and the technical issues of excessively rely on artificial experience during for solving existing driving.
In order to solve the above technical problems, the technical solution adopted in the present invention is as follows：
A kind of that the method for obtaining surrounding rock category is identified based on slag picture, step is as follows：
S1 obtains slag picture.
Slag picture is transmitted to TBM master control rooms and handled by S2, the slag picture that obtains that treated.
S2.1 extracts the chrominance component to take color photographs；
S2.2, gray-scale relation and Gaussian function according to image carry out smooth filter slope；
Image is carried out adaptive histogram balancing, improves the contrast of image by S2.3；
S2.4, the gradient of Multi-Scale Calculation coloured image select optimal value；
S2.5 carries out gray level image Morphological scale-space, the feature of prominent image zooming-out；
S2.6, image carry out extreme value label and watershed segmentation, obtain segmentation image；
S2.7, definition merges and stopping rule carries out region merging technique, more accurately obtains segmentation image；
S2.8 according to area attribute, obtains the characteristic parameter of determinand.
S3, the Grades of Surrounding Rock association sensitive features collection of slag picture after calculation processing.
S3.1 obtains grain size cumulative distribution figure from step S2, and calculates the characteristic value of each slag piece in image.
S3.1.1 calculates nonuniformity coefficient Cu, calculation formula is：
Cu=d60/d10(1)；
Wherein, d60Be limited granulation diameter, is certain grain size in grain size cumulative distribution figure, the native content less than the grain size accounts for total slag
The 60% of piece quantity；
d10- effective grain size, certain grain size in grain size cumulative distribution figure, the native content less than the grain size account for total slag piece quantity
10%.
S3.1.2 calculates coefficient of curvature Cc, calculation formula is：
Wherein：d30For certain grain size in grain size cumulative distribution figure, the native content less than the grain size accounts for total slag piece quantity
30%.
S3.1.3, computational length are more than the sliver quantity N1 of L.
S3.1.4 calculates the slag piece quantity N2 for being less than axis and long axis ratio 0.8.
S3.1.5 calculates axis and is more than or equal to the 0.8 and slag piece quantity N3 less than or equal to 1 with long axis ratio.
S3.1.6 calculates the perimeter C of each slag piece.
S3.1.7 calculates the area S of each slag piece.
S3.1.8 calculates slag piece total quantity N.
S3.2 handles each slag piece characteristic value using ANALYSIS OF RELATIONAL GRADE, obtains Grades of Surrounding Rock association sensitive features collection
[Cu,Cc,N1,N2,N3]。
There are close couplings between the slag piece characteristic value obtained in step S3.1, seriously affect drawing for later stage Grades of Surrounding Rock
Point, therefore correlation analysis method is used, calculate 5 characteristics of image variables and the Grades of Surrounding Rock degree of association.
S4 using obtained surrounding rock category association sensitive features collection as data sample, and uses AP clustering methods by data
Sample divides k cluster；
Specifically calculating step can be：The similarity between data object is first calculated, obtains similar matrix；Then constantly repeatedly
For R and A, cluster center is obtained；Data object is finally divided according to cluster center.
Moreover, AP clustering algorithms are a kind of methods that similarity according between data object is clustered automatically, it is subordinate to
In one kind of partition clustering method.R (m, n) describes the degree that data object n is suitable as the cluster centre of data object m,
What is represented is the message from m to n；A (m, n) describes data object m and data object n is selected to be suitble to as it according to cluster centre
Degree represents the message from n to m.R (m, n) and A (m, n) is bigger, then data object n is more possible as in cluster
The heart.AP clustering algorithms are exactly the Attraction Degree and degree of membership that continuous iteration updates each data object, until certain time of iteration
Remainder data object is assigned in corresponding cluster by number.The formula used is as follows：
(1) Attraction Degree iterative formula
Rt+1(m, n)=(1- λ) Rt+1(m,n)+λ·Rt(m,n)
Wherein：
(2) degree of membership iterative formula：
At+1(m, n)=(1- λ) At+1(m,n)+λ·At(m,n)
Wherein：
S5 by gaussian kernel function and Polynomial kernel function Weighted Fusion, and is LSSVM to each cluster in step S4 and returns
Return, obtain k submodel；
K obtained submodel is weighted fusion, obtains surrounding rock category value, calculation formula is by S6
In formula, μijIt represents that j is under the jurisdiction of the degree of submodel i, is arrived with the distance of sample j to the i-th class center of a sample with sample j
Ratio of all class center of a sample apart from summation represents.
When TBM is tunneled, the slag piece on belt feeder is shot by the High performance industrial camera being erected above belt feeder, is clapped
It acts as regent and puts from face 100m spacing, guarantee analyzes slag piece to obtain in real time.Picture after shooting is reached into TBM master control rooms electricity
Brain is analyzed in real time, obtains treated slag picture.It is sensitive that surrounding rock category association is obtained from treated slag picture
Feature set, and cluster analysis is carried out to surrounding rock category association sensitive features collection and obtains k cluster, then each cluster is carried out
LSSVM is returned, and obtains k submodel, then be weighted fusion to k submodel and obtain surrounding rock category value.The present invention is to real-time
Image is acquired processing, without manually participating in, avoids, excessively by artificial experience, improving the accuracy of fender graded and quick
Property.
Specific embodiment
A kind of that the method for obtaining surrounding rock category is identified based on slag picture, step is as follows：
S1 obtains slag picture.
Slag picture is transmitted to TBM master control rooms and handled by S2, the slag picture that obtains that treated.
S2.1 extracts the chrominance component to take color photographs；
S2.2, gray-scale relation and Gaussian function according to image carry out smooth filter slope；
Image is carried out adaptive histogram balancing, improves the contrast of image by S2.3；
S2.4, the gradient of Multi-Scale Calculation coloured image select optimal value；
S2.5 carries out gray level image Morphological scale-space, the feature of prominent image zooming-out；
S2.6, image carry out extreme value label and watershed segmentation, obtain segmentation image；
S2.7, definition merges and stopping rule carries out region merging technique, more accurately obtains segmentation image；
S2.8 according to area attribute, obtains the characteristic parameter of determinand.
S3, the Grades of Surrounding Rock association sensitive features collection of slag picture after calculation processing.
S3.1 obtains grain size cumulative distribution figure from step S2, and calculates the characteristic value of each slag piece in image.
S3.1.1 calculates nonuniformity coefficient Cu, calculation formula is：
Cu=d60/d10(1)；
Wherein, d60Be limited granulation diameter, is certain grain size in grain size cumulative distribution figure, the native content less than the grain size accounts for total slag
The 60% of piece quantity；
d10- effective grain size, certain grain size in grain size cumulative distribution figure, the native content less than the grain size account for total slag piece quantity
10%.
S3.1.2 calculates coefficient of curvature Cc, calculation formula is：
Wherein：d30For certain grain size in grain size cumulative distribution figure, the native content less than the grain size accounts for total slag piece quantity
30%.
S3.1.3, computational length are more than the sliver quantity N1 of L.
S3.1.4 calculates the slag piece quantity N2 for being less than axis and long axis ratio 0.8.
S3.1.5 calculates axis and is more than or equal to the 0.8 and slag piece quantity N3 less than or equal to 1 with long axis ratio.
S3.1.6 calculates the perimeter C of each slag piece.
S3.1.7 calculates the area S of each slag piece.
S3.1.8 calculates slag piece total quantity N.
S3.2 handles each slag piece characteristic value using ANALYSIS OF RELATIONAL GRADE, obtains Grades of Surrounding Rock association sensitive features collection
[Cu,Cc,N1,N2,N3]。
There are close couplings between the slag piece characteristic value obtained in step S3.1, seriously affect drawing for later stage Grades of Surrounding Rock
Point, therefore correlation analysis method is used, calculate 5 characteristics of image variables and the Grades of Surrounding Rock degree of association.
S4 using obtained surrounding rock category association sensitive features collection as data sample, and uses AP clustering methods by data
Sample divides k cluster；
Specifically calculating step can be：The similarity between data object is first calculated, obtains similar matrix；Then constantly repeatedly
For R and A, cluster center is obtained；Data object is finally divided according to cluster center.
Moreover, AP clustering algorithms are a kind of methods that similarity according between data object is clustered automatically, it is subordinate to
In one kind of partition clustering method.R (m, n) describes the degree that data object n is suitable as the cluster centre of data object m,
What is represented is the message from m to n；A (m, n) describes data object m and data object n is selected to be suitble to as it according to cluster centre
Degree represents the message from n to m.R (m, n) and A (m, n) is bigger, then data object n is more possible as in cluster
The heart.AP clustering algorithms are exactly the Attraction Degree and degree of membership that continuous iteration updates each data object, until certain time of iteration
Remainder data object is assigned in corresponding cluster by number.The formula used is as follows：
(1) Attraction Degree iterative formula
Rt+1(m, n)=(1- λ) Rt+1(m,n)+λ·Rt(m,n)
Wherein：
(2) degree of membership iterative formula：
At+1(m, n)=(1- λ) At+1(m,n)+λ·At(m,n)
Wherein：
S5 by gaussian kernel function and Polynomial kernel function Weighted Fusion, and is LSSVM to each cluster in step S4 and returns
Return, obtain k submodel；
K obtained submodel is weighted fusion, obtains surrounding rock category value, calculation formula is by S6
In formula, μijIt represents that j is under the jurisdiction of the degree of submodel i, is arrived with the distance of sample j to the i-th class center of a sample with sample j
Ratio of all class center of a sample apart from summation represents.
It is illustrated below with specific example.
When TBM is tunneled, the slag piece on belt feeder is shot by the High performance industrial camera being erected above belt feeder, is clapped
It acts as regent and puts from face 100m spacing, guarantee analyzes slag piece to obtain in real time.Picture after shooting is reached into TBM master control rooms electricity
Brain is analyzed in real time, obtains treated slag picture.
Then following process is carried out
1. the selection of Grades of Surrounding Rock slag picture feature
Grades of Surrounding Rock slag piece feature refers to fully react the minimum image character subset of fender graded variation.Further investigate slag
Piece and slag picture are found：For fragmented rock body, other than the pulverized particles under hobboing cutter blade, on belt feeder blocky slag piece compared with
It is more, and slag piece edge is without hobboing cutter impression, be after being shaken when rock mass is tunneled in itself by TBM itself crack increase nature come off shape
Into；For rockmass, crack is less between rock mass, and the formation of slag piece is only influenced, therefore plate by TBM hob knife spacing
Shape slag piece is more；Rock mass between occuping broken and complete, bulk exist with sheet rock mass, and the two ratio is with rock integrity
Degree and change.
From treated, image can obtain grain size cumulative distribution figure, and the feature of slag piece can be obtained in the particle-size accumulation distribution map
It is worth nonuniformity coefficient CuWith coefficient of curvature Cc, two coefficients are defined as follows：
1) nonuniformity coefficient Cu
Cu=d60/d10
In formula：Cu- nonuniformity coefficient；
d60- limitation grain size, certain grain size on particle size distribution curve, the native content less than the grain size account for total slag the piece number
The 60% of amount；
d10- effective grain size, certain grain size on particle size distribution curve, the native content less than the grain size account for total slag the piece number
The 10% of amount；
2) coefficient of curvature Cc
In formula：Cc- nonuniformity coefficient；
d30Certain grain size on-particle size distribution curve, the native content less than the grain size account for the 30% of total slag piece quantity；
The characteristic value of slag piece also has the sliver quantity N1 of big Mr. Yu's length；Less than the quantity N2 of axis and long axis ratio 0.8,
To characterize the sheet of slag piece；Axis and long axis ratio are close to the quantity N3 near 0.8-1, to characterize the bulk of slag piece；
Appearance features obtained by eight slag pictures such as the perimeter C of each slag piece, area S, slag piece total quantity N, but these features it
Between there are close couplings, seriously affected the division of later stage Grades of Surrounding Rock.Therefore, using correlation analysis method, 5 images are calculated
Characteristic variable and the Grades of Surrounding Rock degree of association, it is final to obtain Grades of Surrounding Rock association sensitive features collection [Cu,Cc,N1,N2,N3]。
Moreover, tunneling the starting stage in tunnel TBM, according to the slag piece shot under different surrounding rock rank, and image knowledge is carried out
It does not analyze, obtains the association sensitive features collection [C under each country rock gradeu,Cc, N1, N2, N3], when being tunneled for subsequent tunnel TBM
It accurately and timely carries out fender graded and basic database is provided.
2. it is obtained based on affine multi-model least square method supporting vector machine (LSSVM) hard measurement for propagating (AP) clustering method
Grades of Surrounding Rock
《GB 50218-1994 Standard for classification of engineering rock masses》Middle regulation fender graded is divided into I-V class, five ranks.Consider
The reason of calculating, is represented accordingly by 1-5.It the characteristics of property various informative for rock integrity, is decomposed using multi-model modeling
The thought of synthesis, can be in different surrounding rock integrality adaptive domain, and effective hierarchy system is non-linear, and in each partial model
Accurate approaching to reality surrounding rock category.During modeling, on the basis of surrounding rock category association sensitive features are chosen, all samples are used
Affine propagation (AP) clustering method is classified, and submodel is then established to every one kind, then the result of all submodels is carried out
Synthesis, obtains final result.
Submodel input sample is expressed as：X=[x1,x2,…,xn]T
Wherein：xi=(xi1,xi2,xi3,xi4,xi5), i=1,2 ..., n, xi1,xi2,xi3,xi4,xi5Respectively Cu,Cc,N1,
N2,N3；Submodel output sample is Y=[y1,y2,…,yn]T, yi(i=1,2 ..., n) it is surrounding rock category value.
Hard measurement is as follows：
Step 1. carries out characteristics of image selection after being based on the picture processing of shooting slag, chooses surrounding rock category association sensitive features
Value is as data sample；
Data sample is divided into k cluster by step 2. using AP clustering algorithms；
Gaussian kernel function and Polynomial kernel function Weighted Fusion are done LSSVM recurrence to each cluster respectively, obtained by step 3.
To k submodel；
Step 4. is according to formulaTo k, sub- model-weight fusion exports final surrounding rock category value.
μijIt represents that sample j is under the jurisdiction of the degree of submodel i, is arrived with the distance of sample j to the i-th class center of a sample with sample j
Ratio of all class center of a sample apart from summation represents.
In model in application, in real time acquisition froth images feature and input each submodel obtain submodel output as a result, again
Submodel output result is synthesized to obtain a corresponding country rock grade.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
With within principle, any modification, equivalent replacement, improvement and so on should all be included in the protection scope of the present invention god.
Claims (4)
1. a kind of identify the method for obtaining surrounding rock category based on slag picture, which is characterized in that step is as follows：
S1 obtains slag picture；
Slag picture is transmitted to TBM master control rooms and handled by S2, the slag picture that obtains that treated；
S3, the Grades of Surrounding Rock association sensitive features collection of slag picture after calculation processing；
S4 using obtained surrounding rock category association sensitive features collection as data sample, and uses AP clustering methods by data sample
Divide k cluster；
S5 does LSSVM recurrence by gaussian kernel function and Polynomial kernel function Weighted Fusion, and to each cluster in step S4,
Obtain k submodel；
K obtained submodel is weighted fusion, obtains surrounding rock category value, calculation formula is by S6
In formula, μijRepresent that j is under the jurisdiction of the degree of submodel i, with the distance of sample j to the i-th class center of a sample and sample j to owning
Ratio of the class center of a sample apart from summation represents.
2. according to claim 1 identify the method for obtaining surrounding rock category based on slag picture, which is characterized in that in step
In S2, it is as follows：
S2.1 extracts the chrominance component to take color photographs；
S2.2, gray-scale relation and Gaussian function according to image carry out smooth filter slope；
Image is carried out adaptive histogram balancing, improves the contrast of image by S2.3；
S2.4, the gradient of Multi-Scale Calculation coloured image select optimal value；
S2.5 carries out gray level image Morphological scale-space, the feature of prominent image zooming-out；
S2.6, image carry out extreme value label and watershed segmentation, obtain segmentation image；
S2.7, definition merges and stopping rule carries out region merging technique, more accurately obtains segmentation image；
S2.8 according to area attribute, obtains the characteristic parameter of determinand.
3. according to claim 1 identify the method for obtaining surrounding rock category based on slag picture, which is characterized in that in step
In S3, it is as follows：S3.1 obtains grain size cumulative distribution figure from step S2, and calculates the feature of each slag piece in image
Value；
S3.2 handles each slag piece characteristic value using ANALYSIS OF RELATIONAL GRADE, obtains Grades of Surrounding Rock association sensitive features collection；
There are close couplings between the slag piece characteristic value obtained in step S3.1, seriously affect the division of later stage Grades of Surrounding Rock, because
This uses correlation analysis method, calculates 5 characteristics of image variables and the Grades of Surrounding Rock degree of association.
4. according to claim 3 identify the method for obtaining surrounding rock category based on slag picture, which is characterized in that in step
In S3.1, it is as follows：S3.1.1 calculates nonuniformity coefficient Cu, calculation formula is：
Cu=d60/d10(1)；
Wherein, d60Be limited granulation diameter, is certain grain size in grain size cumulative distribution figure, the native content less than the grain size accounts for total slag the piece number
The 60% of amount；
d10- effective grain size, certain grain size in grain size cumulative distribution figure, the native content less than the grain size account for total slag piece quantity
10%；
S3.1.2 calculates coefficient of curvature Cc, calculation formula is：
Wherein：d30For certain grain size in grain size cumulative distribution figure, the native content less than the grain size accounts for the 30% of total slag piece quantity；
S3.1.3, computational length are more than the sliver quantity N1 of L；
S3.1.4 calculates the slag piece quantity N2 for being less than axis and long axis ratio 0.8；
S3.1.5 calculates axis and is more than or equal to the 0.8 and slag piece quantity N3 less than or equal to 1 with long axis ratio；
S3.1.6 calculates the perimeter C of each slag piece；
S3.1.7 calculates the area S of each slag piece；
S3.1.8 calculates slag piece total quantity N.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810019670.6A CN108182440B (en) | 2018-01-09 | 2018-01-09 | Method for acquiring surrounding rock category based on slag slice image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810019670.6A CN108182440B (en) | 2018-01-09 | 2018-01-09 | Method for acquiring surrounding rock category based on slag slice image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108182440A true CN108182440A (en) | 2018-06-19 |
| CN108182440B CN108182440B (en) | 2020-02-07 |
Family
ID=62550334
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810019670.6A Active CN108182440B (en) | 2018-01-09 | 2018-01-09 | Method for acquiring surrounding rock category based on slag slice image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108182440B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109212164A (en) * | 2018-09-27 | 2019-01-15 | 成都理工大学 | A method of passing through the advanced differentiation tunnel surrounding grade of catalase product |
| CN109358505A (en) * | 2018-10-26 | 2019-02-19 | 中铁工程装备集团有限公司 | A kind of TBM intelligent driving method and system |
| CN109410237A (en) * | 2018-09-14 | 2019-03-01 | 中铁工程装备集团有限公司 | A kind of slag picture dividing method based on laser three-D camera |
| CN109765097A (en) * | 2018-12-21 | 2019-05-17 | 中铁三局集团有限公司 | A kind of tunnel surrounding rapid classification method based on RPD drilling machine |
| CN110378916A (en) * | 2019-07-03 | 2019-10-25 | 浙江大学 | A kind of TBM image based on multitask deep learning is slagged tap dividing method |
| CN110516730A (en) * | 2019-08-20 | 2019-11-29 | 中铁工程装备集团有限公司 | The online stage division of quality of surrounding rock based on PSO-SVM algorithm and image recognition |
| CN111079810A (en) * | 2019-12-06 | 2020-04-28 | 中国铁路设计集团有限公司 | Tunnel surrounding rock grade prediction method based on support vector machine |
| CN113159061A (en) * | 2021-03-10 | 2021-07-23 | 哈尔滨工业大学 | Actual tunnel surrounding rock fragment identification method based on example segmentation |
| CN115408928A (en) * | 2022-07-22 | 2022-11-29 | 北京交通大学 | Surrounding rock parameter prediction method and device based on muck analysis |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103106542A (en) * | 2012-12-13 | 2013-05-15 | 长安大学 | Data analyzing and processing system |
| US20160307312A1 (en) * | 2015-04-15 | 2016-10-20 | Ingrain, Inc. | Method For Determining Fabric And Upscaled Properties Of Geological Sample |
-
2018
- 2018-01-09 CN CN201810019670.6A patent/CN108182440B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103106542A (en) * | 2012-12-13 | 2013-05-15 | 长安大学 | Data analyzing and processing system |
| US20160307312A1 (en) * | 2015-04-15 | 2016-10-20 | Ingrain, Inc. | Method For Determining Fabric And Upscaled Properties Of Geological Sample |
Non-Patent Citations (2)
| Title |
|---|
| ZHAO CHUNLIANG等: "The Data Mining Method Based on Support Vector Machine Applied to Predict Tool Life of TBM", 《2017 29TH CHINESE CONTROL AND DECISION CONFERENCE》 * |
| 吴肖坤等: "基于特征值域的可拓学理论的工程岩体质量评价", 《黄金科学技术》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109410237A (en) * | 2018-09-14 | 2019-03-01 | 中铁工程装备集团有限公司 | A kind of slag picture dividing method based on laser three-D camera |
| CN109212164A (en) * | 2018-09-27 | 2019-01-15 | 成都理工大学 | A method of passing through the advanced differentiation tunnel surrounding grade of catalase product |
| CN109358505A (en) * | 2018-10-26 | 2019-02-19 | 中铁工程装备集团有限公司 | A kind of TBM intelligent driving method and system |
| CN109358505B (en) * | 2018-10-26 | 2022-03-29 | 中铁工程装备集团有限公司 | TBM intelligent driving method and system |
| CN109765097B (en) * | 2018-12-21 | 2021-07-02 | 中铁三局集团有限公司 | Tunnel surrounding rock rapid classification method based on RPD drilling machine |
| CN109765097A (en) * | 2018-12-21 | 2019-05-17 | 中铁三局集团有限公司 | A kind of tunnel surrounding rapid classification method based on RPD drilling machine |
| CN110378916B (en) * | 2019-07-03 | 2021-11-09 | 浙江大学 | TBM image slag segmentation method based on multitask deep learning |
| CN110378916A (en) * | 2019-07-03 | 2019-10-25 | 浙江大学 | A kind of TBM image based on multitask deep learning is slagged tap dividing method |
| CN110516730A (en) * | 2019-08-20 | 2019-11-29 | 中铁工程装备集团有限公司 | The online stage division of quality of surrounding rock based on PSO-SVM algorithm and image recognition |
| CN110516730B (en) * | 2019-08-20 | 2022-09-23 | 中铁工程装备集团有限公司 | Surrounding rock quality online grading method based on PSO-SVM algorithm and image recognition |
| CN111079810A (en) * | 2019-12-06 | 2020-04-28 | 中国铁路设计集团有限公司 | Tunnel surrounding rock grade prediction method based on support vector machine |
| CN113159061A (en) * | 2021-03-10 | 2021-07-23 | 哈尔滨工业大学 | Actual tunnel surrounding rock fragment identification method based on example segmentation |
| CN113159061B (en) * | 2021-03-10 | 2022-02-08 | 哈尔滨工业大学 | Actual tunnel surrounding rock fragment identification method based on example segmentation |
| CN115408928A (en) * | 2022-07-22 | 2022-11-29 | 北京交通大学 | Surrounding rock parameter prediction method and device based on muck analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108182440B (en) | 2020-02-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108182440A (en) | It is a kind of that the method for obtaining surrounding rock category is identified based on slag picture | |
| Meyer et al. | Deep learning based 3d object detection for automotive radar and camera | |
| CN106199557B (en) | A kind of airborne laser radar data vegetation extracting method | |
| CN107392232B (en) | Flotation working condition classification method and system | |
| CN107657279A (en) | A kind of remote sensing target detection method based on a small amount of sample | |
| CN104463199A (en) | Rock fragment size classification method based on multiple features and segmentation recorrection | |
| CN108109139B (en) | Airborne LIDAR three-dimensional building detection method based on gray voxel model | |
| CN103632158A (en) | Forest fire prevention monitor method and forest fire prevention monitor system | |
| CN104049245A (en) | Urban building change detection method based on LiDAR point cloud spatial difference analysis | |
| CN103984953A (en) | Cityscape image semantic segmentation method based on multi-feature fusion and Boosting decision forest | |
| CN107194937A (en) | Tongue image partition method under a kind of open environment | |
| CN104408724A (en) | Depth information method and system for monitoring liquid level and recognizing working condition of foam flotation | |
| CN102867183B (en) | Method and device for detecting littered objects of vehicle and intelligent traffic monitoring system | |
| CN101751679A (en) | Sorting method, detecting method and device of moving object | |
| CN105069395B (en) | Roadmarking automatic identifying method based on Three Dimensional Ground laser scanner technique | |
| CN109215112B (en) | Method for marking single-side point cloud model | |
| CN103514598B (en) | A kind of automatic division method building SFM point cloud | |
| CN105243387A (en) | Open-pit mine typical ground object classification method based on UAV image | |
| CN106778749A (en) | Based on the touring operating area boundary extraction method that concentration class and Delaunay triangles are reconstructed | |
| CN104484410A (en) | Data fusion method and system applied to big data system | |
| Hui et al. | An active learning method for DEM extraction from airborne LiDAR point clouds | |
| CN105469111A (en) | Small sample set object classification method on basis of improved MFA and transfer learning | |
| CN112132014A (en) | Target re-identification method and system based on non-supervised pyramid similarity learning | |
| Emery | Co-simulating total and soluble copper grades in an oxide ore deposit | |
| CN103886335A (en) | Polarized SAR image classifying method based on fuzzy particle swarms and scattering entropy |
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 |