CN107424143A - A kind of mine belt conveyor coal quantity measuring method based on binocular stereo vision depth perception - Google Patents
A kind of mine belt conveyor coal quantity measuring method based on binocular stereo vision depth perception Download PDFInfo
- Publication number
- CN107424143A CN107424143A CN201710239653.9A CN201710239653A CN107424143A CN 107424143 A CN107424143 A CN 107424143A CN 201710239653 A CN201710239653 A CN 201710239653A CN 107424143 A CN107424143 A CN 107424143A
- Authority
- CN
- China
- Prior art keywords
- mrow
- coal charge
- msub
- transport
- coal
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20064—Wavelet transform [DWT]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Battery Electrode And Active Subsutance (AREA)
- Image Analysis (AREA)
Abstract
A kind of mine belt conveyor coal quantity measuring method based on binocular vision depth perception, one velocity sensor of installation gathers travelling belt rate signal and transmitted to host computer in real time on ribbon conveyer, two video cameras being placed in parallel are installed above travelling belt and gather transport coal charge image transmitting in real time to host computer, computer assisted image processing is carried out, is concretely comprised the following steps：Scaling method enhancing transport coal charge image is become using multiresolution wavelet, the only image containing coal charge is partitioned into reference to K means clustering algorithms；Coal charge three-dimensional point cloud information is obtained using binocular vision method；The initial volume of transport coal charge is calculated using Delaunay algorithms, transport coal charge volume is modified with reference to the method for T S fuzzy logic inferences, the detection of coal charge amount is realized using coal weighing formula.The present invention can measure the coal charge amount of current transportation in real time according to the characteristics of transport coal charge surface coal cinder, and error is smaller, validity is high, has practical value, convenient to promote.
Description
Technical field
It is more particularly to a kind of to be based on binocular stereo vision depth the present invention relates to coal mine equipment automatic online detection field
The mine belt conveyor coal quantity measuring method of perception.
Background technology
Coal production is an important indicator for weighing coal mining enterprise's economic benefit, produces the coal detection values in each link
The foundation of output statistics and production management is not only, important material feedback information is also provided for process control and optimization.Currently,
Ribbon conveyer is the equipment that usage amount is most in coal production process, but due to the lack of uniformity of coal mining so that conveying
The colliery freight volume of machine can not keep stable, cause ribbon conveyer to be usually in the non-optimal running status of " low load with strong power ",
Cause a large amount of power consumptions.Therefore, optimizing ribbon conveyer rotating speed according to coal amount turns into the problem of in the urgent need to address, and solves
This problem key and on condition that realizing the detection of belt conveying machine-made egg-shaped or honey-comb coal briquets amount.
At present, belt conveying machine-made egg-shaped or honey-comb coal briquets quantity measuring method has two kinds of belted electronic balance mensuration and laser coal inventory instrument mensuration.
Wherein, belted electronic balance mensuration is widely used, but due to dragging roller of being weighed non-aligned degree, belt tension and running resistance etc.
" belt effect ", it is difficult to correct the factors pair such as weighting error, belt speed error, calibration error, environment influence error in practice
The influence of online coal quantity measurement.Laser coal inventory instrument mensuration is a kind of coal quantity measuring method of full-automatic high precision, using non-contact height
Fast laser measurement mode gathers the three-dimensional information data of transport coal charge, and then obtains coal charge volume and quality.However, present laser
Device is expensive, and popularization utilization cost is too high.
In laser coal inventory instrument, the three-dimensional information of dump is measured by laser.Inspired by this, herein in conjunction with binocular stereo vision
Three-dimensional information e measurement technology realizes the non-contact measurement to belt conveying machine-made egg-shaped or honey-comb coal briquets amount.Binocular stereo vision e measurement technology is by not
Two cameras with position are simultaneously to transport coal charge shooting, according to relative displacement combination phase of the point on coal charge in left images
Answer parameter to calculate three-dimensional coordinate a little, so as to realize the cubing of coal charge, be widely used in miniature workpiece, leaves of plants
In the cubing of hat, poultry, industry spot large scale equipment etc..
Patent publication No. CN105841614A discloses a kind of " ribbon conveyer conveying coal amount visual scanning ranging detection side
Method ", it is true by the single width coal charge graphical analysis to collection with a line source and a CCD camera collection transport coal charge data
Surely coal charge cross-sectional area is transported, conveying capacity of the conveyer belt in time of measuring is obtained through time integral.
Patent publication No. CN101949721 discloses a kind of " coal bunker material level measuring based on laser and binocular vision
And device ", laser beam is produced with a laser, hot spot is formed on coal face；Figure is gathered with the CCD camera of two identical parameters
As right, depth map is calculated to utilization Stereo Matching Algorithm to the image of acquisition, finally realizes coal bunker level gauging.
The document of method and correlation described in above patent make use of the image of camera collection transport coal charge mostly, utilize
The methods of laser, line source, is to draw the depth data of transport coal charge while introduce more noise and being asked using complicated
Topic.In addition, coal charge in actual mine belt conveyor is close with belt color, poor contrast, edge details to be present fuzzy
Problem, cause to transport coal charge identification difficulty, and colliery granular size, shape on the ribbon conveyer in different production procedures
Difference, cause filling rate different so as to transport coal charge there may be different quality under same volume and equal densities and measure
Amount is inaccurate.Notice that above method does not consider above mentioned problem, therefore in the detection of mine belt conveyor transport coal charge amount
It is middle very big error to be present.In summary, a kind of mine belt conveyor coal charge amount based on binocular vision depth perception is proposed
The method of detection is very necessary.
The content of the invention
Goal of the invention：In order to solve the deficiency of existing mine belt conveyor transport coal charge quantity measuring method, for belt
Conveyer transports the characteristics of coal charge, and set forth herein a kind of mine belt conveyor coal charge amount inspection based on binocular vision depth perception
Survey method, detected by transporting coal charge upper horizontal two industrial cameras of installation in mine belt conveyor.
Technical scheme：The present invention proposes coal in a kind of mine belt conveyor based on binocular stereo vision depth perception
Doses detection method, comprises the following steps：
Step 1, identified in the transport coal charge original image through binocular camera collection and coal charge is transported on belt, including：Base
Become the transport coal charge image enhaucament of scaling method in multiresolution wavelet and the transport coal charge image based on K-means clustering algorithms divides
Cut；
Step 2, using the object dimensional data acquisition technology of binocular stereo vision to only containing coal on travelling belt after segmentation
The three-dimensional data of material is extracted, including：Only containing transport coal charge image pair three-dimensional correction, transport the map generalization of coal charge parallax and
The three-dimensional information extraction of coal charge on travelling belt；
Step 3, according to travelling belt coal charge three-dimensional data, calculated by coal charge volume and carry out coal charge amount calculating with correcting,
Including：Transport the calculating of coal charge initial volume, the volume amendment based on T-S fuzzy logic inferences and the calculating for transporting coal charge amount.
Further, in step 1 based on multiresolution wavelet become scaling method transport coal charge image enchancing method include with
Lower four steps：
Step 111, the transport coal charge original image of collection is converted into HSV color spaces by rgb color space, extracts V
The transport coal charge original image of passage；
Step 112, coal charge is transported to the V passages of extraction using multiresolution 2-d discrete wavelet decomposition method such as formula (1)
Original image is decomposed, and obtains transporting the low frequency component and high fdrequency component of coal charge original image V passages；
Wherein, (x, y) is transport coal charge original image pixels coordinate, and (k, l) is the pixel point coordinates decomposed, g and h
Respectively high-pass filter and low pass filter, cjTo transport coal charge original image,It is its vertical high frequency component,It is its horizontal high frequency component,It is its diagonal high fdrequency component, cj+1For its low frequency component；
Step 113, to the low frequency component c of transport coal charge original image V passagesj+1Carry out non-linear contrast's enhancing processing
With the high fdrequency component to V passagesWithCarry out soft-threshold denoising processing；
Step 114, using multiresolution 2-d discrete wavelet reconstructing method to the V multi-channel high frequencies and low frequency component after processing
It is reconstructed such as formula (2), and synthesizes HSV triple channel images, so as to obtains transporting the original enhancing image c of coal chargej(x, y),
The transport coal charge image partition method based on K-means clustering algorithms specifically includes following 5 steps in step 1：
Step 121, in the transport coal charge image basis of enhancing, its number of clusters K=5 included is determined；
Step 122, carry out choosing 5 pixel datas at random in the image of transport coal charge enhancing as transport coal charge image
Split initial cluster center (C1,C2,...,C5)；
Step 123, error sum of squares similarity function such as formula (3) is used to remaining transport according to gray scale and spatial information
Coal charge view data distributes most like cluster；
Wherein, d (xi,xj) it is two pixel x for strengthening coal charge imageiAnd xjThe distance between, σ is one adjustable
Space length scalar, A are the matrixes of adjustable gray scale；piAnd pjRepresent pixel x in coal charge imageiAnd xjNormalization space
Coordinate information vector；siAnd sjRepresent pixel x in coal charge imageiAnd xjNormalized Grey Level information vector；
Step 124, the new cluster centre C in transport coal charge image is recalculatedi；
Step 125, by Optimized Iterative step 123, step 124 finds optimal transport coal charge image clustering, extraction fortune
Class where defeated coal charge, and then obtain the image of the only coal charge containing transport.
Further, step 2 includes following four step：
Step 21, obtain be horizontally arranged at transport coal charge above binocular camera inside and outside portion's parameter include camera focus f,
Principal point coordinate (the c of left and right camerax,cy), distortion factor α, spin matrix R, translation vector T；
Step 22, three-dimensional correction is carried out using the left images of only containing transport coal charge of the camera parameter of acquisition to obtaining；
Step 23, the transport coal charge left images after correction are carried out with Census matching algorithms and draws transport coal charge correction chart
The disparity map of picture；
Step 24, phase is drawn according to the actual parameter of the disparity map of transport coal charge and transport coal charge sampling instrument binocular camera
For the three-dimensional data on the transport coal charge surface of camera.
Further, coal charge initial volume computational methods are transported in step 3 includes：Using three-dimensional Delaunay subdivisions algorithm
Tetrahedron subdivision is carried out to obtaining transport coal charge surface three dimension data through binocular vision, is calculated with tetrahedron volume calculation formula
The transport coal charge initial volume gone out after subdivision；
The volume modification method based on T-S fuzzy logic inferences includes following 4 step in step 3：
Step 321, to the image of the only coal charge containing transport, the edge of transport coal charge surface coal cinder is entered using Canny operators
Row detection, with reference to the method for contours extract, the quantity n of transport coal charge surface coal cinder is calculated, and calculate each coal cinder respectively
Area s and girth l, to transporting the area s and girth l of coal charge surface coal cinder as input variable；
Step 322, input language variable " coal cinder area " and " coal cinder girth " are respectively set to 7 grades：It is { very small
(VS), smaller (SS), small (S), in (M), big (B), bigger (BB), very big (VB) }, coal cinder area and coal cinder girth
Membership function uses Triangleshape grade of membership function；
Step 323, using single order T-S fuzzy logic inferences such as formula (4)：
Wherein, m=49 be coal charge volume T-S fuzzy models in regular number, σiFor the coal charge of i-th rule output
Filling rate, A, B are respectively coal cinder area s and coal cinder girth l fuzzy set, a0、a1、a2For the reality set according to priori
Constant；
Step 324, filling rate σ when transport coal charge is made up of j-th of coal cinder completely can be tried to achieve using weighted mean methodj,
Wherein, σjDetermined by the equation of the i-th rule in T-S fuzzy reasonings, μAiAnd μBiRespectively input coal cinder area and
The membership function of girth, it can obtain transporting approximate filling rate of the coal charge under n different coal cinders based on formula (6)I.e.
And then the volume correction value of coal charge can be calculated
Coal charge amount computational methods are transported in step 3 is：The field range of known camera shooting is L (mm), the speed of belt
For v0, then sampling time Δt=L/v0, in certain time interval T, the coal amount of ribbon conveyer conveying as shown by the equation,
Wherein, M is the coal charge amount that ribbon conveyer conveys in period T,ρFor the coal charge actual density of collection in worksite,
For coal charge volume after the amendment that is calculated in step 324.
Further, the domain of coal cinder area is：{10,30,50,70,90,110,130}；The domain of coal cinder girth is：
{10,20,30,40,50,60,70}。
Beneficial effect：First, a kind of figure being combined based on multiresolution wavelet conversion with K-means clustering methods is proposed
As recognition methods, for the accurate coal charge identification before coal charge binocular vision 3 D information extraction；Secondly, it is proposed that patrolled based on fuzzy
The coal charge volume modification method of reasoning is collected, improves transport coal charge amount computational accuracy.It is analyzed by experiment, institute's extracting method
Satisfied measurement error can be obtained, therefore, the detection of actual industrial ribbon conveyer coal charge amount is can be applied to, is ribbon conveyer
Optimal control effective coal amount real time information is provided.
Brief description of the drawings
Fig. 1 is the coal charge amount detection systems schematic diagram based on binocular vision；
Fig. 2 is the coal quantity measuring method structure based on binocular vision；
Fig. 3 is the coal charge volume correction algorithm structure chart based on T-S fuzzy logic inferences；
Fig. 4 is that T-S fuzzy logics input membership function；
Fig. 5 is the transport original left view of coal charge；
Fig. 6 is transport coal charge original right view；
Fig. 7 is transport coal charge surface three dimension point cloud；
Fig. 8 is the relative error before and after bulk coal feed weight amendment；
Fig. 9 is the relative error before and after medium sized coal material weight amendment；
Figure 10 is the relative error before and after fritter coal feed weight amendment.
Embodiment
The present invention is described in further detail below in conjunction with drawings and the specific embodiments.
Fig. 1 is that the coal charge amount detection systems schematic diagram based on binocular vision installs a velocity pick-up on ribbon conveyer
Device gathers travelling belt rate signal and transmitted to host computer in real time, and two video cameras being placed in parallel are installed above travelling belt
The coal charge image transmitting of collection transport in real time carries out computer assisted image processing to host computer,
The present invention proposes a kind of mine belt conveyor coal charge material detection side based on binocular stereo vision depth perception
Method, comprise the following steps：
Step 1, the transport coal charge on belt is identified from the image comprising transport coal charge collected, including：It is based on
Multiresolution wavelet becomes scaling method transport coal charge image enhaucament and the transport coal charge image segmentation based on K-means clustering algorithms；
Step 2, using the object dimensional cloud data acquiring technology of binocular stereo vision to only containing travelling belt after segmentation
The three-dimensional data of upper coal charge is extracted, including：The only life of the three-dimensional correction of the coal charge image pair containing transport, transport coal charge disparity map
Into the three-dimensional information extraction with coal charge on travelling belt.
Step 3, according to travelling belt coal charge three-dimensional data, calculated by coal charge volume and carry out coal charge amount calculating with correcting,
Including：Transport the calculating of coal charge initial volume, the volume amendment based on T-S fuzzy logic inferences and the calculating for transporting coal charge amount.
In step 1 based on multiresolution wavelet become scaling method transport coal charge image enchancing method be specially：
By the transport coal charge original image of collection by RGB (Red, Green, Blue) color space be converted into HSV (Hue,
Saturation, Value) color space, extracts V channel images, 2-d discrete wavelet decomposition is carried out using formula (1).
Wherein, (x, y) is transport coal charge original image pixels coordinate, and (k, l) is the pixel point coordinates decomposed, g and h
Respectively high-pass filter and low pass filter.cjTo transport coal charge coal original image,It is the vertical of its wavelet decomposition
High fdrequency component,It is the horizontal high frequency component of its wavelet decomposition,It is the diagonal high fdrequency component of its wavelet decomposition,
cj+1For the low frequency component of its wavelet decomposition.Most of coal charge picture noise, coal charge point feature and side as caused by radiation and signal
Edge details etc. is all in high-frequency sub-bandWithIn, and low frequency sub-band cj+1The main profile for characterizing coal charge image etc. is near
Likelihood signal, therefore contrast enhancing is carried out to V channel low frequencies approximation component using nonlinear exponent transform method, and pass through setting
Soft-threshold carries out image denoising processing to V multi-channel high frequencies details coefficients.On this basis, using 2-d discrete wavelet reconstructing method
As formula (2) is reconstructed to the low frequencies of V passages, high fdrequency component, and synthesizes HSV triple channel images, so as to the transport strengthened
Coal charge image.
The transport coal charge image partition method based on K-means clustering algorithms is specially in step 1：
In the transport coal charge image basis of enhancing, point of transport coal charge is realized using non-supervisory K-means clustering methods
Cut and identify.
(1) according to the transport coal charge image of enhancing, its number of clusters K=5 included is determined, classification includes transport coal charge
Class, travelling belt class, other ground noodles, support background classes etc.；
(2) from the transport coal charge image of enhancing, 5 pixel datas are randomly selected as the first of transport coal charge segmentation figure picture
Beginning cluster centre (C1,C2,...,C5)；
(3) error sum of squares similarity function such as formula (3) is used, it is most like that data distribution in coal charge image is transported to residue
Cluster Ci；
Wherein, d (xi,xj) it is two pixel x for strengthening coal charge imageiAnd xjThe distance between, σ is adjustable space
Distance parameter, A are adjustable gray matrixs；piAnd pjRepresent pixel x in coal charge imageiAnd xjNormalization space coordinates letter
Breath vector；siAnd sjRepresent pixel x in coal charge imageiAnd xjNormalized Grey Level information vector.
(4) new cluster centre is recalculated by Euclidean distance formula in the coal charge image of enhancing；
(5) Optimized Iterative (3), (4) step are until new cluster centre is equal with former cluster centre or less than specified threshold, extraction
Transport coal charge where class image with realize only containing transport coal charge image.
The three-dimensional correction method of the only coal charge image pair containing transport includes in step 2:
(1) camera spin matrix R ∈ R are obtained with scaling method3×3With translation vector T ∈ R3×1, and will using formula (4)
Camera spin matrix R is divided into the composite matrix r of left and right cameralAnd rr.Purpose be left and right camera primary optical axis is parallel to each other so that
Re-projection distortion is minimum, realizes that camera is coplanar.
(2) to ensure the alignment of camera row, limit can be transformed into nothing by establishing the row alignment transition matrix of left and right camera
The method of poor distant place is realized, therefore, creating the spin matrix R in translation vector T directions firstrec。
Wherein,For with the equidirectional limits of translation vector T, wherein T=[Tx,Ty,Tz]T, Tx, TyAnd TzRespectively
X, the translation vector in y and z directions；For the vector in plane of delineation direction；e3=e1×e2, it is perpendicular to e1
With e2The vector of place plane.
The R that root above formula calculatesrecWith the composite matrix r of left and right cameralAnd rr, the row of left and right camera can be tried to achieve by formula (5)
It is directed at transition matrix RlAnd Rr。
(3) according to camera imaging model, with reference to the row alignment transition matrix R of left and right cameralAnd Rr, obtained by formula (6)
Transport coal charge left images pixel coordinate after correctionWith
Wherein
MlAnd MrRespectively correct the internal reference matrix of front left and right camera；WithThe internal reference of left and right camera after respectively correcting
Matrix；(ul,vl) and (ur,vr) it is respectively transport coal charge left images pixel coordinate, s before correctinglAnd srRespectively left images
The proportionality coefficient of correction,WithThe focal length of left and right camera in the x direction after respectively correcting,Respectively correct rear left
The focal length of right camera in y-direction.WithThe principal point coordinate of left and right camera after respectively correcting.fxlWith
fxrThe focal length of left and right camera in the x direction after respectively correcting, fylAnd fyrRespectively correct all around that camera is in y-direction
Focal length.(cxl, cyl) and (cxr, cyr) it is respectively the principal point coordinate for correcting front left and right camera.
The generation method of transport coal charge disparity map is specially in step 2:
Using the local matching algorithm converted based on Census, it uses center pixel to be compared with neighborhood territory pixel gray value
Compared with the influence of the factors such as gain and the luminance deviation of image can be reduced, and retains the positional information of neighborhood territory pixel.Algorithm mainly wraps
Include：The Census conversion of coal charge correction chart picture, Matching power flow calculate, cost polymerization, the optimization of coal charge disparity map.
The Census conversion of coal charge correction chart picture is to be by N (p) pixel-maps in topography center pixel P and its neighborhood
One Bit String, then Census conversion codes using this Bit String as center pixel P, so as to replace pixel P original ash
Angle value.The mapping relations of Census conversion are as follows.
Wherein,
ξ(Ip,Iq) represent that pixel q for central pixel point p, passes through its gray value IqAnd IpObtained after comparison operation
Bit.Str (x, y) represents the Census conversion codes of coal charge correction picture centre pixel (x, y)；I (x, y) corrects for coal charge
The gray value of picture centre pixel (x, y)；I (x+i, y+j) denotation coordination is the pixel gray value of (x+i, y+j)；L is window
The half of mouth width, r are the half of window height；Accorded with for bit serial.
After Census conversion is carried out to coal charge correction chart picture, obtained using Hamming distance similarity measure function (9)
Matching power flow.
C (x, y, d)=Hamming (Strl(x,y),Strr(x-d,y)) (9)
Wherein, Strl(x, y) and Strr(x-d, y) be respectively coal charge correction left image and right image in coordinate be (x, y) as
The Census conversion codes of vegetarian refreshments；C (x, y, d) is Matching power flow value, is pixel (x, y) in left image and right image
The different number in position of Census conversion codes and the difference of corresponding pixel points abscissa in d expression left and right coal charge correction chart pictures.
According to Matching power flow value C (x, y, d), using fixed-size rectangular window, the polymerization cost in it is calculated.
Wherein, m and n is respectively the half of polymerizing windows width and height.
According to the polymerization cost C ' calculatedsum, in disparity range (dmin,dmax) in pass through the method formula that the victor is a king
(11) obtain each pixel point coordinates in coal charge correction chart picture and polymerize cost argminC (x, y, d) for the minimum of (x, y), so as to
Initial parallax D (x, y) corresponding to obtaining.
Hereafter, the Mismatching point in transport coal charge initial parallax figure is determined using left and right consistency constraint formula (12).
|dRL(x+dLR(x,y),y)-dLR(x, y) | ＜ σ (12)
Wherein, dLR(x, y) is that coal charge corrects parallax of the right figure relative to left figure midpoint (x, y)；dRL(x+dLR(x,y),y)
Left figure midpoint (x, y), which is corrected, for coal charge corresponds to right figure midpoint (x+dLR(x, y), y) parallax；It is 1 to set σ, it is determined that transport coal
Expect the Mismatching point in initial parallax figure.To eliminate the Mismatching point transported in coal charge initial parallax figure, further using cross
Intersect substitution value of 4 parallax averages as Mismatching point up and down in neighborhood, so as to Optimizing Transport coal charge initial parallax
Figure, draws dense transport coal charge disparity map.
The 3-D information fetching method of coal charge is specially on travelling belt in step 2:
The transport coal charge disparity map obtained by the sectional perspective matching algorithm converted based on Census, with reference to by Zhang Zhengyou
The camera parameter that scaling method obtains, recover the three-dimensional point of transport coal charge with the principle of triangulation such as formula (13) of binocular vision
Cloud information.
Wherein, (X, Y, Z) is coordinate of the transport coal charge surface point relative to left camera, and (x, y) is transportation table millet cake on a left side
The coordinate of the plane of delineation, TxFor the translation vector in x directions,Left principal point for camera coordinate after being corrected for binocular, fxFor x side
To equivalent focal length.Therefore, the three-dimensional point cloud information for transporting coal charge surface point can be by the transport coal that is shot by left and right camera
Material left images recover.
Coal charge initial volume computational methods are transported in step 3 is specially:
To the transport coal charge surface three dimension discrete point cloud data obtained through binocular stereo vision, calculated using three-dimensional Delaunay
Empty the circumsphere criterion and local criterion of method, carry out subdivision.
First, the cloud data of coal charge is traveled through, obtains the enveloped box of coal charge point cloud, a convex closure is built, in the convex of generation
Initial tetrahedral mesh is built in bag, then, anchor point is inputted in the initial tetrahedral mesh of foundation, will as new summit
Tetrahedron subdivision updates tetrahedral grid, until being all inserted into tetrahedron a little in coal charge point cloud into four new tetrahedrons
Grid, finally, optimize the tetrahedron of generation by LOP (Local Optimization Procedure) local algorithm criterion.
All tetrahedrons generated by coal charge cloud data are traveled through, formula (14) obtains each tetrahedron volume,
Wherein, (x1,y1,z1)、(x2,y2,z2)、(x3,y3,z3) and (x4,y4,z4) it is tetrahedral four apex coordinates.
All tetrahedron volumes are summed, so as to calculate the initial volume of coal charge.
The volume modification method based on T-S fuzzy logic inferences is as follows in step 3:
Because ribbon conveyer coal charge is in block random distribution, cause to produce non-uniform gap inside coal charge, and gap is big
It is small not known with random, and Delaunay algorithms are that coal charge volume is calculated under gapless assumed condition inside coal charge,
Therefore larger error be present.To obtain more accurate coal charge volume, identify that the shape of coal charge surface coal cinder is big first herein
It is small, and then T-S fuzzy logic inference methods are combined, decision-making goes out suitable coal charge filling rate, so as to reduce coal charge gap to volume
Influence, improve volume computational accuracy.Coal charge volume correction algorithm based on T-S fuzzy logic inferences is as shown in figure 3, specific mistake
Journey is as described below.
1) to the only image containing coal charge, the edge of coal charge surface coal cinder is detected using Canny operators, with reference to profile
Extracting method, calculates the quantity n of coal charge surface coal cinder, and calculates the area s and girth l of each coal cinder respectively, to be used as T-S
The input variable of fuzzy logic.
2) according to many experiments result, input language variable " coal cinder area " and " coal cinder girth " are respectively set to 7
Grade：Very small (VS), smaller (SS), small (S), in (M), big (B), bigger (BB), very big (VB) }；Coal cinder area
Domain be：{10,30,50,70,90,110,130}；The domain of coal cinder girth is：{10,20,30,40,50,60,70}.Coal
The membership function of block area and coal cinder girth uses Triangleshape grade of membership function.
3) the coal charge volume amendment of T-S fuzzy reasonings
The amendment of coal charge volume uses single order T-S fuzzy logic inferences, as shown in formula (15).
Wherein, m=49, for number regular in the T-S fuzzy models of coal charge volume；σiFor the coal of i-th rule output
Expect filling rate；A, B is respectively coal cinder area s and coal cinder girth l fuzzy set；a0、a1、a2Set according to priori
Real constant.
The filling rate transported when coal charge is made up of j-th of coal cinder completely can be tried to achieve using weighted mean method according to formula (16)
σj。
Wherein, σiDetermined by the equation of the i-th rule in T-S fuzzy reasonings, μAiAnd μBiRespectively input coal cinder area and
The membership function of girth.It can obtain transporting approximate filling rate of the coal charge under n different coal cinders based on formula (16)I.e.
And then the volume correction value of coal charge can be calculated
Transport the calculating of coal charge amount：
If the field range of camera shooting is L (mm), the belt speed gathered through velocity sensor is v0, then sampling time
Δ t=L/v0.In certain time interval T, shown in the coal charge amount M such as formula (18) of ribbon conveyer conveying.
Wherein, ρ is the coal charge actual density of collection in worksite.
Described above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should
It is considered as protection scope of the present invention.
Claims (5)
1. coal charge quantity measuring method in a kind of mine belt conveyor based on binocular stereo vision depth perception, its feature exist
In comprising the following steps：
Step 1, identified in the transport coal charge original image through binocular camera collection and coal charge is transported on belt, including：Based on more
Resolution wavelet becomes the transport coal charge image enhaucament of scaling method and the transport coal charge image segmentation based on K-means clustering algorithms；
Step 2, using binocular stereo vision object dimensional data acquisition technology to after segmentation only containing coal charge on travelling belt
Three-dimensional data is extracted, including：The only three-dimensional correction of the coal charge image pair containing transport, transport coal charge parallax map generalization and transport
The three-dimensional information extraction of coal charge on belt；
Step 3, according to travelling belt coal charge three-dimensional data, calculated by coal charge volume and carry out coal charge amount calculating with correcting, including：
Transport the calculating of coal charge initial volume, the volume amendment based on T-S fuzzy logic inferences and the calculating for transporting coal charge amount.
2. coal charge amount detects in the mine belt conveyor according to claim 1 based on binocular stereo vision depth perception
Method, it is characterised in that become the transport coal charge image enchancing method of scaling method in step 1 based on multiresolution wavelet including following
Four steps：
Step 111, the transport coal charge original image of collection is converted into HSV color spaces by rgb color space, extracts V passages
Transport coal charge original image；
Step 112, it is original to the V passages transport coal charge of extraction using multiresolution 2-d discrete wavelet decomposition method such as formula (1)
Image is decomposed, and obtains transporting the low frequency component and high fdrequency component of coal charge original image V passages；
<mrow>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<msubsup>
<mi>d</mi>
<mrow>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mi>V</mi>
</msubsup>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munder>
<mo>&Sigma;</mo>
<mi>k</mi>
</munder>
<munder>
<mo>&Sigma;</mo>
<mi>l</mi>
</munder>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>-</mo>
<mn>2</mn>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mi>h</mi>
<mrow>
<mo>(</mo>
<mi>l</mi>
<mo>-</mo>
<mn>2</mn>
<mi>y</mi>
<mo>)</mo>
</mrow>
<msub>
<mi>c</mi>
<mi>j</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>,</mo>
<mi>l</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<msubsup>
<mi>d</mi>
<mrow>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mi>H</mi>
</msubsup>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munder>
<mo>&Sigma;</mo>
<mi>k</mi>
</munder>
<munder>
<mo>&Sigma;</mo>
<mi>l</mi>
</munder>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>-</mo>
<mn>2</mn>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mi>h</mi>
<mrow>
<mo>(</mo>
<mi>l</mi>
<mo>-</mo>
<mn>2</mn>
<mi>y</mi>
<mo>)</mo>
</mrow>
<msub>
<mi>c</mi>
<mi>j</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>,</mo>
<mi>l</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<msubsup>
<mi>d</mi>
<mrow>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mi>D</mi>
</msubsup>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munder>
<mo>&Sigma;</mo>
<mi>k</mi>
</munder>
<munder>
<mo>&Sigma;</mo>
<mi>l</mi>
</munder>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>-</mo>
<mn>2</mn>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mi>h</mi>
<mrow>
<mo>(</mo>
<mi>l</mi>
<mo>-</mo>
<mn>2</mn>
<mi>y</mi>
<mo>)</mo>
</mrow>
<msub>
<mi>c</mi>
<mi>j</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>,</mo>
<mi>l</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<msub>
<mi>c</mi>
<mrow>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munder>
<mo>&Sigma;</mo>
<mi>k</mi>
</munder>
<munder>
<mo>&Sigma;</mo>
<mi>l</mi>
</munder>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>-</mo>
<mn>2</mn>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mi>h</mi>
<mrow>
<mo>(</mo>
<mi>l</mi>
<mo>-</mo>
<mn>2</mn>
<mi>y</mi>
<mo>)</mo>
</mrow>
<msub>
<mi>c</mi>
<mi>j</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>,</mo>
<mi>l</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein, (x, y) is transport coal charge original image pixels coordinate, and (k, l) is the pixel point coordinates decomposed, g and h difference
For high-pass filter and low pass filter, cjTo transport coal charge original image,It is its vertical high frequency component,
It is its horizontal high frequency component,It is its diagonal high fdrequency component, cj+1For its low frequency component；
Step 113, to the low frequency component c of transport coal charge original image V passagesj+1Carrying out non-linear contrast strengthens processing and to V
The high fdrequency component of passageWithCarry out soft-threshold denoising processing；
Step 114, the V multi-channel high frequencies after processing and low frequency component are carried out using multiresolution 2-d discrete wavelet reconstructing method
Reconstruct such as formula (2), and synthesizes HSV triple channel images, so as to obtain transporting the original enhancing image c of coal chargej(x, y),
<mrow>
<mtable>
<mtr>
<mtd>
<mrow>
<msub>
<mi>c</mi>
<mi>j</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mo>=</mo>
<munder>
<mo>&Sigma;</mo>
<mi>l</mi>
</munder>
<munder>
<mo>&Sigma;</mo>
<mi>k</mi>
</munder>
<mi>h</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>-</mo>
<mn>2</mn>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mi>h</mi>
<mrow>
<mo>(</mo>
<mi>l</mi>
<mo>-</mo>
<mn>2</mn>
<mi>y</mi>
<mo>)</mo>
</mrow>
<msub>
<mi>c</mi>
<mrow>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>,</mo>
<mi>l</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mo>+</mo>
<munder>
<mo>&Sigma;</mo>
<mi>l</mi>
</munder>
<munder>
<mo>&Sigma;</mo>
<mi>k</mi>
</munder>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>-</mo>
<mn>2</mn>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mi>h</mi>
<mrow>
<mo>(</mo>
<mi>l</mi>
<mo>-</mo>
<mn>2</mn>
<mi>y</mi>
<mo>)</mo>
</mrow>
<msubsup>
<mi>d</mi>
<mrow>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mi>H</mi>
</msubsup>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>,</mo>
<mi>l</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mo>+</mo>
<munder>
<mo>&Sigma;</mo>
<mi>l</mi>
</munder>
<munder>
<mo>&Sigma;</mo>
<mi>k</mi>
</munder>
<mi>h</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>-</mo>
<mn>2</mn>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>l</mi>
<mo>-</mo>
<mn>2</mn>
<mi>y</mi>
<mo>)</mo>
</mrow>
<msubsup>
<mi>d</mi>
<mrow>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mi>V</mi>
</msubsup>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>,</mo>
<mi>l</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mo>+</mo>
<munder>
<mo>&Sigma;</mo>
<mi>l</mi>
</munder>
<munder>
<mo>&Sigma;</mo>
<mi>k</mi>
</munder>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>-</mo>
<mn>2</mn>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>l</mi>
<mo>-</mo>
<mn>2</mn>
<mi>y</mi>
<mo>)</mo>
</mrow>
<msubsup>
<mi>d</mi>
<mrow>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mi>D</mi>
</msubsup>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>,</mo>
<mi>l</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
</mtable>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</mrow>
The transport coal charge image partition method based on K-means clustering algorithms specifically includes following 5 steps in step 1：
Step 121, in the transport coal charge image basis of enhancing, its number of clusters K=5 included is determined；
Step 122, carry out choosing 5 pixel datas at random in the image of transport coal charge enhancing as the image segmentation of transport coal charge
Initial cluster center (C1,C2,...,C5)；
Step 123, error sum of squares similarity function such as formula (3) is used to remaining transport coal charge according to gray scale and spatial information
View data distributes most like cluster；
<mrow>
<mtable>
<mtr>
<mtd>
<mrow>
<mi>d</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>,</mo>
<msub>
<mi>x</mi>
<mi>j</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>K</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>p</mi>
<mi>i</mi>
</msub>
<mo>,</mo>
<msub>
<mi>p</mi>
<mi>j</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mo>=</mo>
<mi>exp</mi>
<mrow>
<mo>(</mo>
<mo>-</mo>
<mfrac>
<mrow>
<mo>|</mo>
<mo>|</mo>
<msub>
<mi>p</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<msub>
<mi>p</mi>
<mi>j</mi>
</msub>
<mo>|</mo>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
</mrow>
<mi>&sigma;</mi>
</mfrac>
<mo>)</mo>
</mrow>
<mo>&CenterDot;</mo>
<mi>exp</mi>
<mrow>
<mo>(</mo>
<mo>-</mo>
<msup>
<mrow>
<mo>(</mo>
<mrow>
<msub>
<mi>s</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<msub>
<mi>s</mi>
<mi>j</mi>
</msub>
</mrow>
<mo>)</mo>
</mrow>
<mi>T</mi>
</msup>
<mi>A</mi>
<mo>(</mo>
<mrow>
<msub>
<mi>s</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<msub>
<mi>s</mi>
<mi>j</mi>
</msub>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
</mtable>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>3</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein, d (xi,xj) it is two pixel x for strengthening coal charge imageiAnd xjThe distance between, σ be an adjustable space away from
From scalar, A is the matrix of adjustable gray scale；piAnd pjRepresent pixel x in coal charge imageiAnd xjNormalization space coordinates letter
Breath vector；siAnd sjRepresent pixel x in coal charge imageiAnd xjNormalized Grey Level information vector；
Step 124, the new cluster centre C in transport coal charge image is recalculatedi；
Step 125, by Optimized Iterative step 123, step 124 finds optimal transport coal charge image clustering, extraction transport coal
Class where expecting, and then obtain the image of the only coal charge containing transport.
3. coal charge amount detects in the mine belt conveyor according to claim 1 based on binocular stereo vision depth perception
Method, it is characterised in that step 2 includes following four step：
Step 21, obtain be horizontally arranged at transport coal charge above binocular camera inside and outside portion's parameter include camera focus f, left and right
Principal point coordinate (the c of camerax,cy), distortion factor α, spin matrix R, translation vector T；
Step 22, three-dimensional correction is carried out using the left images of only containing transport coal charge of the camera parameter of acquisition to obtaining；
Step 23, to after correction transport coal charge left images carry out Census matching algorithms draw transport coal charge correction chart as
Disparity map；
Step 24, according to transport coal charge disparity map and transport coal charge sampling instrument binocular camera actual parameter draw relative to
The three-dimensional data on the transport coal charge surface of camera.
4. coal charge amount detects in the mine belt conveyor according to claim 1 based on binocular stereo vision depth perception
Method, it is characterised in that
Coal charge initial volume computational methods are transported in step 3 to be included：Using three-dimensional Delaunay subdivisions algorithm to through binocular vision
Obtain transport coal charge surface three dimension data and carry out Tetrahedron subdivision, the fortune after subdivision is calculated with tetrahedron volume calculation formula
Defeated coal charge initial volume；
The volume modification method based on T-S fuzzy logic inferences includes following 4 step in step 3：
Step 321, to the image of the only coal charge containing transport, the edge of transport coal charge surface coal cinder is examined using Canny operators
Survey, with reference to the method for contours extract, calculate the quantity n of transport coal charge surface coal cinder, and calculate the area s of each coal cinder respectively
With girth l, to the area s and girth l of transport coal charge surface coal cinder as input variable；
Step 322, input language variable " coal cinder area " and " coal cinder girth " are respectively set to 7 grades：It is { very small
(VS), smaller (SS), small (S), in (M), big (B), bigger (BB), very big (VB) }, coal cinder area and coal cinder girth
Membership function uses Triangleshape grade of membership function；
Step 323, using single order T-S fuzzy logic inferences such as formula (4)：
<mrow>
<mtable>
<mtr>
<mtd>
<mrow>
<msub>
<mi>R</mi>
<mi>i</mi>
</msub>
<mo>:</mo>
<mi>i</mi>
<mi>f</mi>
<mi> </mi>
<mi>s</mi>
<mo>=</mo>
<mi>A</mi>
<mi> </mi>
<mi>a</mi>
<mi>n</mi>
<mi>d</mi>
<mi> </mi>
<mi>l</mi>
<mo>=</mo>
<mi>B</mi>
<mo>,</mo>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mtable>
<mtr>
<mtd>
<mrow>
<mi>t</mi>
<mi>h</mi>
<mi>e</mi>
<mi>n</mi>
</mrow>
</mtd>
<mtd>
<mrow>
<msub>
<mi>&sigma;</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<msubsup>
<mi>a</mi>
<mn>0</mn>
<mi>i</mi>
</msubsup>
<mo>+</mo>
<msubsup>
<mi>a</mi>
<mn>1</mn>
<mi>i</mi>
</msubsup>
<mi>s</mi>
<mo>+</mo>
<msubsup>
<mi>a</mi>
<mn>2</mn>
<mi>i</mi>
</msubsup>
<mi>l</mi>
</mrow>
</mtd>
</mtr>
</mtable>
<mo>,</mo>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
<mo>,</mo>
<mn>2</mn>
<mo>,</mo>
<mn>...</mn>
<mo>,</mo>
<mi>m</mi>
</mrow>
</mtd>
</mtr>
</mtable>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>4</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein, m=49 be coal charge volume T-S fuzzy models in regular number, σiFor the coal charge filling of i-th rule output
Rate, A, B are respectively coal cinder area s and coal cinder girth l fuzzy set, a0、a1、a2For the real constant set according to priori；
Step 324, filling rate σ when transport coal charge is made up of j-th of coal cinder completely can be tried to achieve using weighted mean methodj,
<mrow>
<msub>
<mi>&sigma;</mi>
<mi>j</mi>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>m</mi>
</munderover>
<msub>
<mi>&mu;</mi>
<mrow>
<mi>A</mi>
<mi>i</mi>
</mrow>
</msub>
<msub>
<mi>&mu;</mi>
<mrow>
<mi>B</mi>
<mi>i</mi>
</mrow>
</msub>
<msub>
<mi>&sigma;</mi>
<mi>i</mi>
</msub>
</mrow>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>m</mi>
</munderover>
<msub>
<mi>&mu;</mi>
<mrow>
<mi>A</mi>
<mi>i</mi>
</mrow>
</msub>
<msub>
<mi>&mu;</mi>
<mrow>
<mi>B</mi>
<mi>i</mi>
</mrow>
</msub>
</mrow>
</mfrac>
<mo>,</mo>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
<mo>,</mo>
<mn>2</mn>
<mo>,</mo>
<mo>...</mo>
<mo>,</mo>
<mi>n</mi>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>5</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein, σjDetermined by the equation of the i-th rule in T-S fuzzy reasonings, μAiAnd μBiRespectively input coal cinder area and girth
Membership function, can obtain transporting approximate filling rate of the coal charge under n different coal cinders based on formula (6)I.e.
<mrow>
<mover>
<mi>&sigma;</mi>
<mo>&OverBar;</mo>
</mover>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<mi>n</mi>
</mfrac>
<munderover>
<mo>&Sigma;</mo>
<mi>j</mi>
<mi>n</mi>
</munderover>
<msub>
<mi>&sigma;</mi>
<mi>j</mi>
</msub>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>6</mn>
<mo>)</mo>
</mrow>
</mrow>
And then the volume correction value of coal charge can be calculated
Coal charge amount computational methods are transported in step 3 is：The field range of known camera shooting is L (mm), and the speed of belt is v0,
Then sampling time Δ t=L/v0, in certain time interval T, the coal amount of ribbon conveyer conveying as shown by the equation,
<mrow>
<mtable>
<mtr>
<mtd>
<mrow>
<mi>M</mi>
<mo>=</mo>
<mi>&rho;</mi>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>m</mi>
</munderover>
<msub>
<mover>
<mi>V</mi>
<mo>&OverBar;</mo>
</mover>
<mi>i</mi>
</msub>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mi>m</mi>
<mo>=</mo>
<mi>T</mi>
<mo>/</mo>
<mi>&Delta;</mi>
<mi>t</mi>
</mrow>
</mtd>
</mtr>
</mtable>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>7</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein, M is the coal charge amount that ribbon conveyer conveys in period T, and ρ is the coal charge actual density of collection in worksite,For step
Coal charge volume after the amendment being calculated in rapid 324.
5. coal charge amount detects in the mine belt conveyor according to claim 4 based on binocular stereo vision depth perception
Method, it is characterised in that；The domain of coal cinder area is：{10,30,50,70,90,110,130}；The domain of coal cinder girth is：
{10,20,30,40,50,60,70}。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710239653.9A CN107424143B (en) | 2017-04-13 | 2017-04-13 | Mining belt conveyor coal quantity detection method based on binocular stereoscopic vision depth perception |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710239653.9A CN107424143B (en) | 2017-04-13 | 2017-04-13 | Mining belt conveyor coal quantity detection method based on binocular stereoscopic vision depth perception |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107424143A true CN107424143A (en) | 2017-12-01 |
| CN107424143B CN107424143B (en) | 2020-10-16 |
Family
ID=60423304
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710239653.9A Active CN107424143B (en) | 2017-04-13 | 2017-04-13 | Mining belt conveyor coal quantity detection method based on binocular stereoscopic vision depth perception |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107424143B (en) |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108225502A (en) * | 2017-12-06 | 2018-06-29 | 中南大学 | A kind of truck loads ore quality method of estimation and system |
| CN108267210A (en) * | 2017-12-21 | 2018-07-10 | 北方重工装备（沈阳）有限公司 | A kind of shield machine is unearthed measurement and early warning system |
| CN108317974A (en) * | 2018-04-09 | 2018-07-24 | 广州华润热电有限公司 | One discharge plate coal method and disk coal measures system |
| CN108510539A (en) * | 2018-05-09 | 2018-09-07 | 中国矿业大学(北京) | A kind of coal production monitoring method based on image analysis |
| CN108629776A (en) * | 2018-05-16 | 2018-10-09 | 郑州大学 | Ore-rock granularity Detection system |
| CN109163775A (en) * | 2018-08-30 | 2019-01-08 | 北京广天夏科技有限公司 | A kind of mass measuring method and device based on ribbon conveyer |
| CN109341823A (en) * | 2018-11-15 | 2019-02-15 | 力博重工科技股份有限公司 | Stream real-time detection apparatus based on video image |
| CN109655019A (en) * | 2018-10-29 | 2019-04-19 | 北方工业大学 | Cargo volume measurement method based on deep learning and three-dimensional reconstruction |
| CN110246079A (en) * | 2019-05-23 | 2019-09-17 | 上海交通大学 | Camera distortion antidote, system and medium based on B-spline surface fitting |
| CN110260801A (en) * | 2019-05-13 | 2019-09-20 | 平安科技（深圳）有限公司 | Method and apparatus for measuring volume of material |
| CN110675443A (en) * | 2019-09-24 | 2020-01-10 | 西安科技大学 | Coal briquette area detection method for underground coal conveying image |
| KR102088137B1 (en) * | 2019-09-05 | 2020-03-11 | 변형완 | Device and method for selecting humate from lump coal |
| CN111553850A (en) * | 2020-03-30 | 2020-08-18 | 深圳一清创新科技有限公司 | Three-dimensional information acquisition method and device based on binocular stereo vision |
| CN111638159A (en) * | 2020-05-29 | 2020-09-08 | 新兴铸管股份有限公司 | Blast furnace raw fuel particle size detection method |
| CN111768331A (en) * | 2020-06-24 | 2020-10-13 | 深兰人工智能芯片研究院（江苏）有限公司 | Optimization method and device |
| CN112046957A (en) * | 2019-06-05 | 2020-12-08 | 西安瑞德宝尔智能科技有限公司 | Method and device for monitoring and processing ore blocking |
| CN112330693A (en) * | 2020-11-13 | 2021-02-05 | 北京伟景智能科技有限公司 | Coal gangue detection method and system |
| CN112561887A (en) * | 2020-12-18 | 2021-03-26 | 中国矿业大学 | Belt conveyor coal flow binocular vision measurement method based on deep migration learning |
| CN112632793A (en) * | 2020-12-30 | 2021-04-09 | 中国矿业大学 | Off-line calculation method for coal bunker coal material bulk density |
| CN112700419A (en) * | 2020-12-31 | 2021-04-23 | 精英数智科技股份有限公司 | Yield measuring method, device and system based on image segmentation |
| CN112910167A (en) * | 2020-11-02 | 2021-06-04 | 泰州程顺制冷设备有限公司 | Motor speed reduction system applying parameter analysis |
| CN113800223A (en) * | 2021-10-11 | 2021-12-17 | 中国神华能源股份有限公司哈尔乌素露天煤矿 | Method, device and system for detecting coal conveying amount of belt conveyor |
| CN113984156A (en) * | 2021-09-22 | 2022-01-28 | 北京伟景智能科技有限公司 | Belt weighing method and device based on optical measurement and storage medium |
| CN114021983A (en) * | 2021-11-05 | 2022-02-08 | 宁夏广天夏电子科技有限公司 | Three-dimensional running monitoring method and system |
| CN114463357A (en) * | 2022-01-27 | 2022-05-10 | 枣庄矿业(集团)付村煤业有限公司 | Method for determining dynamic information of medium pile in real time in dense medium coal preparation |
| CN115641455A (en) * | 2022-09-16 | 2023-01-24 | 杭州视图智航科技有限公司 | Image matching method based on multi-feature fusion |
| CN116750436A (en) * | 2023-08-15 | 2023-09-15 | 四川蜀道建筑科技有限公司 | But height automatically regulated's feeding belt feeder |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103226005A (en) * | 2013-03-22 | 2013-07-31 | 中国计量学院 | System and method for measuring weight of stacked objects by mesh-surface structure light |
| CN103364057A (en) * | 2013-07-11 | 2013-10-23 | 赛摩电气股份有限公司 | Belt transport coal dump weight and quality detection device |
| US20140049616A1 (en) * | 2005-05-10 | 2014-02-20 | Roger Stettner | Dimensioning system |
| CN104132699A (en) * | 2014-07-24 | 2014-11-05 | 武汉中原电子集团有限公司 | Laser scanning type bulk material flow detection and distribution error elimination method |
-
2017
- 2017-04-13 CN CN201710239653.9A patent/CN107424143B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140049616A1 (en) * | 2005-05-10 | 2014-02-20 | Roger Stettner | Dimensioning system |
| CN103226005A (en) * | 2013-03-22 | 2013-07-31 | 中国计量学院 | System and method for measuring weight of stacked objects by mesh-surface structure light |
| CN103364057A (en) * | 2013-07-11 | 2013-10-23 | 赛摩电气股份有限公司 | Belt transport coal dump weight and quality detection device |
| CN104132699A (en) * | 2014-07-24 | 2014-11-05 | 武汉中原电子集团有限公司 | Laser scanning type bulk material flow detection and distribution error elimination method |
Non-Patent Citations (5)
| Title |
|---|
| 王俊孟: "基于双目立体视觉的传送带上煤的动态体积测量研究", 《中国优秀硕士学位论文全文数据库 工程科技I辑》 * |
| 苏亚坤等: "绪论", 《基于T-S模糊模型的非线性系统的控制与滤波设计》 * |
| 苏士美等: "骨髓细胞图像的小波变换与K_means聚类分割算法", 《郑州大学学报（工学版）》 * |
| 许晓丽: "基于聚类分析的图像分割算法研究", 《中国博士学位论文全文数据库信息科技辑》 * |
| 金汉均 等: "小波变换在HSV颜色空间上的图像检索应用研究", 《电子测量技术》 * |
Cited By (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108225502A (en) * | 2017-12-06 | 2018-06-29 | 中南大学 | A kind of truck loads ore quality method of estimation and system |
| CN108267210A (en) * | 2017-12-21 | 2018-07-10 | 北方重工装备（沈阳）有限公司 | A kind of shield machine is unearthed measurement and early warning system |
| CN108317974A (en) * | 2018-04-09 | 2018-07-24 | 广州华润热电有限公司 | One discharge plate coal method and disk coal measures system |
| CN108510539A (en) * | 2018-05-09 | 2018-09-07 | 中国矿业大学(北京) | A kind of coal production monitoring method based on image analysis |
| CN108629776B (en) * | 2018-05-16 | 2022-02-08 | 郑州大学 | Mineral rock granularity detecting system |
| CN108629776A (en) * | 2018-05-16 | 2018-10-09 | 郑州大学 | Ore-rock granularity Detection system |
| CN109163775A (en) * | 2018-08-30 | 2019-01-08 | 北京广天夏科技有限公司 | A kind of mass measuring method and device based on ribbon conveyer |
| CN109655019A (en) * | 2018-10-29 | 2019-04-19 | 北方工业大学 | Cargo volume measurement method based on deep learning and three-dimensional reconstruction |
| CN109341823A (en) * | 2018-11-15 | 2019-02-15 | 力博重工科技股份有限公司 | Stream real-time detection apparatus based on video image |
| CN110260801A (en) * | 2019-05-13 | 2019-09-20 | 平安科技（深圳）有限公司 | Method and apparatus for measuring volume of material |
| CN110246079B (en) * | 2019-05-23 | 2020-11-06 | 上海交通大学 | B-spline surface fitting-based camera distortion correction method, system and medium |
| CN110246079A (en) * | 2019-05-23 | 2019-09-17 | 上海交通大学 | Camera distortion antidote, system and medium based on B-spline surface fitting |
| CN112046957A (en) * | 2019-06-05 | 2020-12-08 | 西安瑞德宝尔智能科技有限公司 | Method and device for monitoring and processing ore blocking |
| KR102088137B1 (en) * | 2019-09-05 | 2020-03-11 | 변형완 | Device and method for selecting humate from lump coal |
| CN110675443A (en) * | 2019-09-24 | 2020-01-10 | 西安科技大学 | Coal briquette area detection method for underground coal conveying image |
| CN110675443B (en) * | 2019-09-24 | 2022-12-20 | 西安科技大学 | Coal briquette area detection method for underground coal conveying image |
| CN111553850A (en) * | 2020-03-30 | 2020-08-18 | 深圳一清创新科技有限公司 | Three-dimensional information acquisition method and device based on binocular stereo vision |
| CN111553850B (en) * | 2020-03-30 | 2023-10-24 | 深圳一清创新科技有限公司 | Three-dimensional information acquisition method and device based on binocular stereoscopic vision |
| CN111638159A (en) * | 2020-05-29 | 2020-09-08 | 新兴铸管股份有限公司 | Blast furnace raw fuel particle size detection method |
| CN111768331A (en) * | 2020-06-24 | 2020-10-13 | 深兰人工智能芯片研究院（江苏）有限公司 | Optimization method and device |
| CN112910167A (en) * | 2020-11-02 | 2021-06-04 | 泰州程顺制冷设备有限公司 | Motor speed reduction system applying parameter analysis |
| CN112330693A (en) * | 2020-11-13 | 2021-02-05 | 北京伟景智能科技有限公司 | Coal gangue detection method and system |
| CN112330693B (en) * | 2020-11-13 | 2023-12-29 | 北京伟景智能科技有限公司 | Gangue detection method and system |
| CN112561887A (en) * | 2020-12-18 | 2021-03-26 | 中国矿业大学 | Belt conveyor coal flow binocular vision measurement method based on deep migration learning |
| CN112561887B (en) * | 2020-12-18 | 2021-06-29 | 中国矿业大学 | Belt conveyor coal flow binocular vision measurement method based on deep migration learning |
| CN112632793B (en) * | 2020-12-30 | 2021-11-02 | 中国矿业大学 | Off-line calculation method for coal bunker coal material bulk density |
| CN112632793A (en) * | 2020-12-30 | 2021-04-09 | 中国矿业大学 | Off-line calculation method for coal bunker coal material bulk density |
| CN112700419A (en) * | 2020-12-31 | 2021-04-23 | 精英数智科技股份有限公司 | Yield measuring method, device and system based on image segmentation |
| CN113984156A (en) * | 2021-09-22 | 2022-01-28 | 北京伟景智能科技有限公司 | Belt weighing method and device based on optical measurement and storage medium |
| CN113800223A (en) * | 2021-10-11 | 2021-12-17 | 中国神华能源股份有限公司哈尔乌素露天煤矿 | Method, device and system for detecting coal conveying amount of belt conveyor |
| CN114021983A (en) * | 2021-11-05 | 2022-02-08 | 宁夏广天夏电子科技有限公司 | Three-dimensional running monitoring method and system |
| CN114463357A (en) * | 2022-01-27 | 2022-05-10 | 枣庄矿业(集团)付村煤业有限公司 | Method for determining dynamic information of medium pile in real time in dense medium coal preparation |
| CN114463357B (en) * | 2022-01-27 | 2024-05-24 | 枣庄矿业(集团)付村煤业有限公司 | Method for determining dynamic information of medium pile in real time in dense medium coal dressing |
| CN115641455A (en) * | 2022-09-16 | 2023-01-24 | 杭州视图智航科技有限公司 | Image matching method based on multi-feature fusion |
| CN115641455B (en) * | 2022-09-16 | 2024-01-09 | 杭州视图智航科技有限公司 | Image matching method based on multi-feature fusion |
| CN116750436A (en) * | 2023-08-15 | 2023-09-15 | 四川蜀道建筑科技有限公司 | But height automatically regulated's feeding belt feeder |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107424143B (en) | 2020-10-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107424143A (en) | A kind of mine belt conveyor coal quantity measuring method based on binocular stereo vision depth perception | |
| CN115272341B (en) | Packaging machine defect product detection method based on machine vision | |
| CN105184801B (en) | It is a kind of based on multi-level tactful optics and SAR image high-precision method for registering | |
| CN101915913B (en) | Steady automatic matching method for high-resolution satellite image connecting points | |
| CN103258214B (en) | Based on the Classifying Method in Remote Sensing Image of image block Active Learning | |
| CN111047695B (en) | Method for extracting height spatial information and contour line of urban group | |
| CN110032949A (en) | A kind of target detection and localization method based on lightweight convolutional neural networks | |
| CN104820991B (en) | A kind of multiple soft-constraint solid matching method based on cost matrix | |
| CN112561887B (en) | Belt conveyor coal flow binocular vision measurement method based on deep migration learning | |
| CN109141248A (en) | Pig weight measuring method and system based on image | |
| CN105279772B (en) | A kind of trackability method of discrimination of infrared sequence image | |
| CN106530337A (en) | Non local stereopair dense matching method based on image gray scale guiding | |
| CN110533774B (en) | Three-dimensional model reconstruction method based on smart phone | |
| CN107314957A (en) | A kind of measuring method of rock fragmentation Size Distribution | |
| CN112950532B (en) | Train pantograph state detection method | |
| CN112435267B (en) | Disparity map calculation method for high-resolution urban satellite stereo image | |
| CN103834796B (en) | A kind of method of strip running deviation in ONLINE RECOGNITION stove | |
| CN106355646B (en) | Surface thinning method, device and system applied to image procossing | |
| CN112116553B (en) | Passive three-dimensional point cloud model defect identification method based on K-D tree | |
| CN110502977A (en) | A kind of building classification of change detection method, system, device and storage medium | |
| CN112017234A (en) | Stockpile volume measurement method based on sparse point cloud reconstruction | |
| CN115880586A (en) | Satellite remote sensing image cloud and snow detection method based on mixed feature network | |
| CN115063472A (en) | Deep learning-based luggage identification and measurement method and device | |
| CN110926426A (en) | Real-time detection device and detection method for residual muck volume of subway shield segment | |
| CN113963054A (en) | Intelligent express box size measuring method and system based on binocular camera |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |