CN106709905A - Vibration-proof hammer fault online detection and identification method based on binocular vision image - Google Patents
Vibration-proof hammer fault online detection and identification method based on binocular vision image Download PDFInfo
- Publication number
- CN106709905A CN106709905A CN201611116535.0A CN201611116535A CN106709905A CN 106709905 A CN106709905 A CN 106709905A CN 201611116535 A CN201611116535 A CN 201611116535A CN 106709905 A CN106709905 A CN 106709905A
- Authority
- CN
- China
- Prior art keywords
- damper
- image
- failure
- thing
- fixed reference
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
- G06T2207/30184—Infrastructure
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a vibration-proof hammer fault online detection and identification method based on a binocular vision image. The method comprises the following steps of using two cameras to form a binocular stereo vision system; establishing one vibration-proof hammer state attribute set S; under an original state, acquiring initial image information possessing a vibration-proof hammer and a fixed reference object, through the initial image information, acquiring a three-dimensional coordinate of the vibration-proof hammer, a three-dimensional coordinate of the fixed reference object and then acquiring an initial value S0 of the vibration-proof hammer state attribute set S; under a current state, when the vibration-proof hammer is discovered to have an abnormal fault, acquiring current image information possessing the vibration-proof hammer and the fixed reference object, through the current image information, acquiring the three-dimensional coordinate of the vibration-proof hammer, the three-dimensional coordinate of the fixed reference object and then acquiring a current value St of the vibration-proof hammer state attribute set S; identifying a vibration-proof hammer fault type; and according to the current value St and the initial value S0, calculating a changing value delta S of a vibration-proof hammer state attribute, and according to the delta S, determining and acquiring a vibration-proof hammer fault condition.
Description
Technical field
The present invention relates to the detection technique of damper, and in particular to a kind of damper failure based on binocular vision image exists
Line detection recognition method.
Background technology
The effect of the meteorological conditions such as overhead transmission line wind-engaging, ice, low temperature, makes circuit produce vibration and wave, if shaken
Dynamic frequency is higher and amplitude very little, and wind shake is dynamic to make trolley wire at suspension point repeatedly by stubborn folding, can cause fatigue of materials, finally leads
Cause stranded, line-outage contingency;If the frequency waved is very low, and amplitude is very big, it is easy to causes alternate flashover, causes circuit to jump
The major accidents such as lock, power failure or burn wire.
Damper is an indispensable part in overhead transmission line safety measure, be important vibrationproof gold utensil it
One, it is carried additionally at circuit tower bar suspension point, can be absorbed or damping vibration energy, changes circuit wobble frequency, prevents circuit
Violent oscillatory motion and wave, be that the life-span of transmission line of electricity and the safety in transmission of electricity corridor play very important effect.Accordingly, with respect to
The research of damper is very meaningful.
At present, the detection recognition method to damper failure is also considerably less, and also neither one method be capable of identify that it is anti-
The algorithm of all fault types of vibration hammer.But in reality, damper is usually present displacement failure, damages failure, release failure, different
The failures such as thing interference, these failures can threaten to the safety in transmission line of electricity life-span and transmission of electricity corridor.
The content of the invention
The technical problems to be solved by the invention are to solve the fault detect to stockbridge damper, it is therefore intended that provide one kind and be based on
The damper On-line Fault Detection recognition methods of binocular vision image.
The present invention is achieved through the following technical solutions:
A kind of damper On-line Fault Detection recognition methods based on binocular vision image, comprises the following steps:
Using two image mechanisms into Binocular Stereo Vision System;
Set up a damper status attribute set S;
In an initial condition, the initial image information that there is damper, fixed reference according to thing is obtained using two video cameras,
Belonged to so as to obtain damper state according to thing three-dimensional coordinate by the acquisition damper three-dimensional coordinate of initial image information, fixed reference
The initial value S of property set S0;
Under current state, the abnormal failure detection of damper when finding that abnormal failure occurs in damper, is taken the photograph using two
Camera obtains the present image information for having damper, fixed reference according to thing, three-dimensional by present image information acquisition damper
Coordinate, fixed reference are according to thing three-dimensional coordinate so as to obtain the currency S of damper status attribute set St;
Damper fault type recognition:According to currency StWith initial value S0Calculate the changing value △ of damper status attribute
S, judges to obtain the situation of damper failure according to △ S.
Design principle of the invention is:The present invention obtains image based on Binocular Stereo Vision System and binocular vision system
The method of three-dimensional coordinate is carried out, and the present invention also sets up damper status attribute set S based on damper most common failure, according to double
Item stereo vision system and image three-dimensional sit damper status attribute in calibration method measuring and calculating practical work process, by these vibration hammers
Status attribute gathers to form set S, shaking when vibration hammer status attribute during using original state, i.e. fault-free is with failure
Hammer status attribute is compared, and vibration hammer fault message is analyzed from comparison result.Its General layout Plan is divided into four-stage:
First stage is:With predetermined damper status attribute set S, then under by original state, using binocular stereo vision
System and acquisition image three-dimensional sit calibration method to obtain initial damper three-dimensional coordinate, fixed reference according to thing three-dimensional coordinate,
We can be deduced in the case where knowing damper three-dimensional coordinate, fixed reference according to thing three-dimensional coordinate damper state and
Damper and fixed reference under the conditions of so, can combine to form damper status attribute collection according to the spatial relationship between thing
Close the initial value S of S0.Second stage is:Whether damper enters into event under judging current state using abnormal conditions monitoring technology
Barrier state.Phase III is:Nonserviceable in the case of confirmation, use utilize Binocular Stereo Vision System and acquisition again
Image three-dimensional sits calibration method to obtain current damper three-dimensional coordinate, fixed reference according to thing three-dimensional coordinate, likewise, we
Damper state and damper can be deduced in the case of knowing damper three-dimensional coordinate, fixed reference according to thing three-dimensional coordinate
With fixed reference according to the spatial relationship between thing, under the conditions of so, can combine to form working as damper status attribute set S
Preceding value St.Fourth stage is:Compare currency StWith initial value S0Between situation of change, analyzed according to these situations of change
Malfunction.
Generally speaking, the technical scheme is that using Binocular Stereo Vision System, i.e. dual camera technique for taking, obtaining
The steric parameter of object is obtained, steric parameter analysis association attributes is recycled, and by the change of front and rear attribute so as to draw event
Barrier state.Designing points of the invention are whole method processing procedure, are analyzed by such steric parameter, and we can obtain
To many fault parameters, so as to analyze damper malfunction comprehensively, for example, the steric parameter by analyzing damper, just
Can directly obtain under original state and under current state damper structure change, it is also possible to by the solid for analyzing damper
Parameter and fixed reference analyze fixed reference according to the distance between thing and damper, it is also possible to analyze according to the steric parameter of thing
Go out the number change of damper, it is necessary to when designing certain attribute of detection damper, I designs coherent detection according to the actual requirements
Model, these cannot not design a model fixedly and select particular model, are defined by that can calculate fault message using steric parameter, this
The central scope of invention aims to provide a kind of method and obtains damper steric parameter, and the design of fault message is analyzed from steric parameter
Scheme, above-mentioned steric information is to refer to the three-dimensional coordinate information obtained in such scheme.
Preferably, because damper is usually present displacement failure, damage failure, release failure, foreign matter interference failure, we
The damper status attribute of design is related above-mentioned failure, is specifically expressed as follows:The damper status attribute set S includes
ι、m、r、n;ι represents damper with fixed reference according to the distance between thing (can represent displacement failure);M represents a damper
Iron hammer quantity, under normal circumstances, damper has two iron hammers (can represent damage failure);R represents chaff interference to anti-
The distance (can represent that foreign matter disturbs failure) of vibration hammer;N represents that a certain section of damper number of overhead route (can represent the event that comes off
Barrier);Then initial value S0Including ι0、m0、r0、n0;Currency StIncluding ιt、mt、rt、nt;The changing value △ S include △ ι, △ m,
△ r, △ n, wherein, △ ι=ιt- ι0, △ m=mt- m0, △ r=rt- r0, △ n=nt- n0。ι0Prevent under expression original state
Vibration hammer is with fixed reference according to the distance between thing, m0Represent iron hammer quantity, the r of the next damper of original state0Represent initial shape
Minimum standard value, the n of the distance of chaff interference and damper under state0The damper number of overhead route under expression original state;ιt
Damper shines the distance between thing, m with fixed reference under representing current statetRepresent the iron hammer of the next damper of current state
Quantity, rtDistance, the n of chaff interference and damper under expression current statetThe damper number of overhead route under expression current state.
Wherein, because damper displacement failure is difficult to judge, therefore, one fixed reference of present invention design shines thing, with admittedly
The fixed change with reference to according to the distance between thing and damper is recognized between damper with the presence or absence of displacement, in order to know
Do not go out the shift amount of damper, our preferred fixed references are overhead route wire clamp according to thing, due to overhead route wire clamp and damper all
It is arranged on same thread guide path, therefore is advantageously provided with and identifies shift amount, it is also possible to fixed reference is set to according to thing
Any damper or one in addition the reference substance of setting, other objects that can also be on tower bar, can also be tower bar in itself.Together
When, damper has two iron hammers under normal circumstances, and in the case of thunderbolt or other, iron hammer occurs that damage is dropped so that vibrationproof
Hammer is uneven, therefore, it can obtain depth map by the three-dimensional coordinate of damper, from depth map analysis iron hammer quantity, by preceding
Comparative analysis afterwards has and damages failure, therefore, damaging failure can be represented by detecting iron hammer number change.Further, since anti-
Vibration hammer drop failure occur when, the quantity of damper can change, therefore the number of damper can be judged by 3 D stereo parameter
Amount, represents that damper drops failure by the number change of damper.Due to requiring with damper center with r0In radius
Chaff interference must not occur, therefore, when chaff interference has been detected, we are not chaff interferences again by 3 D stereo parameter
Steric information, chaff interference calculate between damper by the steric information of chaff interference and the steric information of damper
Distance, so as to detect chaff interference failure, wherein r0Belong in advance according to the value set about requirement, by rtWith r0Difference
Value represents chaff interference failure.
Include following process into Binocular Stereo Vision System using two image mechanisms:
Build binocular stereo vision hardware system:Two video cameras are fixed on optical table with the spacing of parallax range B
On;
Shoot scaling board image sets:Make longitude and latitude case marker fixed board complete imaging in two cameras, shoot multigroup different attitudes
Under longitude and latitude case marker fixed board image;
Obtain parameter:Coordinate system with any camera extracts longitude and latitude case marker fixed board image angle point, with most as world coordinate system
Small square law obtains the distortion parameter of the inner parameter matrix, external parameter matrix and left and right camera lens of left and right camera.
The abnormal failure detection of damper includes following process:
Obtain background image:Turn off video camera A, the image obtained by video camera B is used as background image;
Abnormality detection:Current first two field picture is shot by video camera B, abnormality detection is carried out using Background difference, there is different
Often turn fault detect during detection;
Fault detect:Current second two field picture is shot by video camera B, fault detect, background are carried out using Background difference
The time of image, the time of current first two field picture was earlier than current second frame figure earlier than the time of current first two field picture
The time of picture.The image for wherein being obtained by video camera B can be to turn off video camera A as the moment lights, be obtained by video camera B
The first two field picture for obtaining.
Carrying out abnormality detection using Background difference includes following process:
Current first two field picture I that video camera B is shott(x, y) and background image ut(x, y) does subtraction and obtains
Background difference;Background difference is Dt(x, y)=| It(x,y)-ut(x,y)|;Work as DtDuring (x, y) >=ε, then it is judged to exception occur,
ε is unusual determination threshold value;
Carrying out fault detect using Background difference includes following process:
Obtain current second two field picture I of video camera B shootingst+1(x, y) and background image ut(x, y) does subtraction and obtains
To fault detect background difference;Fault detect background difference is Dt+1(x, y)=| It+1(x, y)-ut(x, y) |;Work as Dt+1(x,y)
During >=ε, then it is judged to break down, ε is unusual determination threshold value.
By it is above-mentioned as can be seen that the present invention in the abnormal failure detection process of damper, employ abnormality detection and
Two processes of fault detect, its reason is, because detection environment is outside exploitation formula environment, can usually there is bird, leaf etc. miscellaneous
Thing is by camera visual range, but the appearance process of these debris is very of short duration, therefore, if only carrying out exception successively
Detection, then can trigger many wrong reports, and the operation of current failure analysis is just opened under the only 2 times regulations of detection of the present invention, keep away
Exempt from system and be in frequent starting state, whole system is standby under can maintaining low power consumpting state before the fault detect, also will not
There is multiple wrong report, reduce line walking number of times, current second two field picture in the present invention refers to that current first is later than on the time
The image of individual two field picture.In order to improve the ageing of detection and prevent false triggering fault alarm, current second frame of the invention
Next two field picture of current first two field picture of image selection, if for example, we are by current second two field picture and current
It is long that the time interval of one two field picture is set, it is assumed that bird process occurs in current first two field picture, it is possible to can be current
There is leaf in second two field picture, this will also result in false triggering fault alarm, therefore, only by current second two field picture with
The time interval of current first two field picture set it is smaller when, can just judge that the object of moment passes through, to prevent wrong report therefore
Barrier, so as to improve the ageing and accuracy of detection and accuracy of detection.
It is described below down, the present invention obtains the process of three-dimensional coordinate:
Obtain damper three-dimensional coordinate, fixed reference by initial image information includes following process according to thing three-dimensional coordinate:
In an initial condition, left image, right image are obtained by left video camera, right video camera respectively, forms initial pictures
Information;
Left image, right image are carried out to eliminate distortion and polar curve correction process, left image and right figure after making elimination distort
As strictly row is corresponding;
Image preprocessing is carried out to left image and right image;
Left image and right image characteristic point are extracted respectively;
Matching left image and the characteristic point of right image, obtain multigroup matching double points;
Obtain image three-dimensional coordinate:Left image is obtained with right image in respective pixel using binocular stereo vision measurement model
The three-dimensional coordinate of coordinate system;
Realize that pixel coordinate is tied to the conversion of world coordinate system:The characteristic point of left image is set in the pixel coordinate system of oneself
Pixel coordinate (xl, yl), the characteristic point of right image the pixel coordinate system of oneself pixel coordinate (xr, yr), (xl, yl) with
(xr, yr) matching double points are formed, in the space orientation of image, image measurement is distant, a minor variations of pixel coordinate
Also huge measurement error can be caused.Present invention selection is extracted subpixel coordinates and forms matching again after the completion of image sequence Point matching
Point is right so that the three-dimensional measuring result of image is more accurate, image scene object more accurate positioning.
Three dimensional space coordinate (X of the matching double points in world coordinate systemW,YW,ZW) relation be shown below:
Wherein, xlAnd xrRepresent that left image, right image obtain horizontal seat of the matching double points under pixel coordinate system after matching
Mark, ylAnd yrRepresent that left image, right image obtain ordinate of the matching double points under pixel coordinate system after matching, B represents that a left side is taken the photograph
Parallax range between camera and right video camera, f represents the focal length of video camera where world coordinate system, and B and f are obtained according to camera calibration
Arrive;
If matching double points are the matching double points of damper, three dimensional space coordinate of the damper in world coordinate system
(X0(stockbridge), Y0(stockbridge), Z0(stockbridge)) relation be shown below:
If matching double points be fixed reference according to thing matching double points, fixed reference according to thing world coordinate system three-dimensional space
Between coordinate (X0(clamp), Y0(clamp), Z0(clamp)) relation be shown below:
Obtain the initial value S of damper status attribute set S0Including following process:
Obtain l0:Detect that damper and fixed reference shine thing, according to damper, fixed reference according to thing in world coordinate system
Three dimensional space coordinate calculate damper and fixed reference according to the distance between thing l0, wherein,
Obtain m0:Directly configure m0=2 obtain depth map and obtain 3D point cloud data again by the three-dimensional coordinate of damper
The iron hammer number m of damper is detected by 3D matching process0, wherein m0Represent an iron hammer quantity for damper;
Obtain r0:Directly configure r0Value, r0Represent chaff interference to the distance of damper, r now0It is damper center
Radius;
Obtain n0:The electric wire in image is detected, window is then found with a damper and is slided along electric wire, often detected
To a damper, damper quantity adds 1, final n0Represent a certain section of damper number of overhead route;
ι0、m0、r0、n0It is the initial value S of damper status attribute set S0。
Following process is included according to thing three-dimensional coordinate by present image information acquisition damper three-dimensional coordinate, fixed reference:
Under current state, left image, right image are obtained by left video camera, right video camera respectively, form present image
Information;
Left image, right image are carried out to eliminate distortion and polar curve correction process, left image and right figure after making elimination distort
As strictly row is corresponding;
Image preprocessing is carried out to left image and right image;
Left image and right image characteristic point are extracted respectively;
Matching left image and the characteristic point of right image, obtain multigroup matching double points;
Obtain image three-dimensional coordinate:Left image is obtained with right image in respective pixel using binocular stereo vision measurement model
The three-dimensional coordinate of coordinate system;
Realize that pixel coordinate is tied to the conversion of world coordinate system:The characteristic point of left image is set in the pixel coordinate system of oneself
Pixel coordinate (xl, yl), the characteristic point of right image the pixel coordinate system of oneself pixel coordinate (xr, yr), (xl, yl) with
(xr, yr) form matching double points, three dimensional space coordinate (X of the matching double points in world coordinate systemW,YW,ZW) relation such as following formula institute
Show:
Wherein, xlAnd xrRepresent that left image, right image obtain horizontal seat of the matching double points under pixel coordinate system after matching
Mark, ylAnd yrRepresent that left image, right image obtain ordinate of the matching double points under pixel coordinate system after matching, B represents that a left side is taken the photograph
Parallax range between camera and right video camera, f represents the focal length of video camera where world coordinate system, and B and f are obtained according to camera calibration
Arrive;
If matching double points are the matching double points of damper, three dimensional space coordinate of the damper in world coordinate system
(Xt(stockbridge), Yt(stockbridge), Zt(stockbridge)) relation be shown below:
If matching double points be fixed reference according to thing matching double points, fixed reference according to thing world coordinate system three-dimensional space
Between coordinate (Xt(clamp), Yt(clamp), Zt(clamp)) relation be shown below:
Whether identification present image information has damper chaff interference, if with the presence of damper chaff interference, also including obtaining
Take damper chaff interference three-dimensional coordinate process as follows:
The matching double points that the matching double points are damper chaff interference are recognized, then damper chaff interference is in world coordinate system
Three dimensional space coordinate (Xt(distractors), Yt(distractors), Zt(distractors)) relation be shown below:
Obtain the currency S of damper status attribute set StIncluding following process:
Obtain lt:Detect that damper and fixed reference shine thing, according to damper, fixed reference according to thing in world coordinate system
Three dimensional space coordinate calculate damper and fixed reference according to the distance between thing lt, wherein
Obtain mt:3D point cloud data are obtained by the three-dimensional coordinate acquisition depth map of damper to be detected by 3D matching process again
Go out the iron hammer number m of dampert, wherein mtRepresent an iron hammer quantity for damper;
Obtain rt:
If centered on damper, with r0Have to be identified present image information using frame difference method in the range of radius
Damper chaff interference, then calculate damper chaff interference to damper apart from rt, r0It is the value of directly configuration, wherein,
If centered on damper, with r0Do not have to be identified present image information using frame difference method in the range of radius
Damper chaff interference, then assign damper chaff interference to damper apart from rt, assign rt=r0;
Obtain nt:The electric wire in image is detected, window is then found with a damper and is slided along electric wire, often detected
To a damper, damper quantity adds 1, final ntRepresent a certain section of damper number of overhead route;
lt、mt、rt、ntIt is the currency S of damper status attribute set St。
Judge that the situation for obtaining damper failure includes procedure below according to △ S:
Output displacement failure:If Δ l ∈ [a, b], then illustrate that damper, in the absence of displacement failure, otherwise, has displacement event
Barrier, wherein, a, b is respectively the lower and upper limit of damper and fixed reference according to the safe distance scope between thing, according to △ ι's
Value represents damper translocation distance;
Output iron hammer damages failure:If Δ m=0, show damper in the absence of damaged condition, otherwise, damper is deposited
Failure is being damaged, the value according to △ m represents that iron hammer damages quantity;
Output damper chaff interference failure:If Δ r≤0, illustrate that damper has foreign matter interference failure, according to △ r
Value represent chaff interference to the distance of damper, otherwise in the absence of damper chaff interference failure;
Output damper release failure:If Δ n ≠ 0, illustrate there is damper release failure, the value table according to △ n
Show the quantity that damper comes off, otherwise in the absence of damper release failure.
The characteristics of property according to damper each fault category, can be judged obtain the feelings of failure by state changing features amount
Condition.It is specifically described as:
1) the distance between damper and overhead route wire clamp have certain safety standard, if this distance is not in safety
Within standard, then it is assumed that be to be shifted over, that is, there is displacement failure.I.e.:If Δ l ∈ [a, b], then illustrate that damper does not exist
, otherwise, there is displacement failure in displacement failure.Wherein, a, b are respectively the safe distance scope between damper and overhead route wire clamp
Lower and upper limit.
2) damper is made up of a bit of metal of a geometrical clamp and two with two iron hammers, and two must have two iron
Hammer, if only one of which or do not had, referred to as damages.Therefore the RM for damaging failure is:
If Δ m=0, show damper in the absence of damaged condition, otherwise, damper has damage failure.
3) in the certain limit around damper, do not allow the presence of other object, otherwise damper is interfered,
I.e. there is foreign matter interference failure in damper.Damper foreign matter interference Fault Identification mode be:If Δ r≤0, illustrates damper
There is foreign matter interference failure, do not exist otherwise.
4) need damper quantity to be mounted more than one in some overhead routes, if the quantity for detecting is not standard
Damper quantity, then must be that some dampers have completely fallen off.Damper release failure decision method is:
If Δ n ≠ 0, illustrate there is damper release failure, do not exist otherwise.
Damper fault type list is finally exported, and failure concrete condition is exported.For example, a=0.01 meters, b=
0.05 meter, if damper malfunction property set is Δ S={ 0.03, -1, -0.5,0 }, then corresponding output result is:It is anti-
There is damage failure and foreign matter interference failure in vibration hammer.If Δ S={ 3,0,0,0 }, then illustrate that damper has displacement failure, and
And shift value is 3 meters.
The present invention compared with prior art, has the following advantages and advantages:
1. the state of damper is described by setting up the set of a damper status attribute, by damper failure numerical value
Change, make the fault detect of damper convenient, fast, directly perceived, unified, comprehensive.
2. damper image three-dimensional coordinate is set up using Binocular Stereo Vision System, the damper size of acquisition is accurate.
3. all values of damper status attribute set are detected using binocular stereo vision system, can accurately be surveyed
Value.
Specific embodiment
To make the object, technical solutions and advantages of the present invention become more apparent, with reference to embodiment, the present invention is made
Further to describe in detail, exemplary embodiment of the invention and its explanation are only used for explaining the present invention, are not intended as to this
The restriction of invention.
Embodiment
A kind of damper On-line Fault Detection recognition methods based on binocular vision image, comprises the following steps:
Using two image mechanisms into Binocular Stereo Vision System;
Set up a damper status attribute set S;
In an initial condition, the initial image information that there is damper, fixed reference according to thing is obtained using two video cameras,
Belonged to so as to obtain damper state according to thing three-dimensional coordinate by the acquisition damper three-dimensional coordinate of initial image information, fixed reference
The initial value S of property set S0;
Under current state, the abnormal failure detection of damper when finding that abnormal failure occurs in damper, is taken the photograph using two
Camera obtains the present image information for having damper, fixed reference according to thing, three-dimensional by present image information acquisition damper
Coordinate, fixed reference are according to thing three-dimensional coordinate so as to obtain the currency S of damper status attribute set St;
Damper fault type recognition:According to currency StWith initial value S0Calculate the changing value △ of damper status attribute
S, judges to obtain the situation of damper failure according to △ S.
Design principle of the invention is:The present invention obtains image based on Binocular Stereo Vision System and binocular vision system
The method of three-dimensional coordinate is carried out, and the present invention also sets up damper status attribute set S based on damper most common failure, according to double
Item stereo vision system and image three-dimensional sit damper status attribute in calibration method measuring and calculating practical work process, by these vibration hammers
Status attribute gathers to form set S, shaking when vibration hammer status attribute during using original state, i.e. fault-free is with failure
Hammer status attribute is compared, and vibration hammer fault message is analyzed from comparison result.Its General layout Plan is divided into four-stage:
First stage is:With predetermined damper status attribute set S, then under by original state, using binocular stereo vision
System and acquisition image three-dimensional sit calibration method to obtain initial damper three-dimensional coordinate, fixed reference according to thing three-dimensional coordinate,
We can be deduced in the case where knowing damper three-dimensional coordinate, fixed reference according to thing three-dimensional coordinate damper state and
Damper and fixed reference under the conditions of so, can combine to form damper status attribute collection according to the spatial relationship between thing
Close the initial value S of S0.Second stage is:Whether damper enters into event under judging current state using abnormal conditions monitoring technology
Barrier state.Phase III is:Nonserviceable in the case of confirmation, use utilize Binocular Stereo Vision System and acquisition again
Image three-dimensional sits calibration method to obtain current damper three-dimensional coordinate, fixed reference according to thing three-dimensional coordinate, likewise, we
Damper state and damper can be deduced in the case of knowing damper three-dimensional coordinate, fixed reference according to thing three-dimensional coordinate
With fixed reference according to the spatial relationship between thing, under the conditions of so, can combine to form working as damper status attribute set S
Preceding value St.Fourth stage is:Compare currency StWith initial value S0Between situation of change, analyzed according to these situations of change
Malfunction.
Generally speaking, the technical scheme is that using Binocular Stereo Vision System, i.e. dual camera technique for taking, obtaining
The steric parameter of object is obtained, steric parameter analysis association attributes is recycled, and by the change of front and rear attribute so as to draw event
Barrier state.Designing points of the invention are whole method processing procedure, are analyzed by such steric parameter, and we can obtain
To many fault parameters, so as to analyze damper malfunction comprehensively, for example, the steric parameter by analyzing damper, just
Can directly obtain under original state and under current state damper structure change, it is also possible to by the solid for analyzing damper
Parameter and fixed reference analyze fixed reference according to the distance between thing and damper, it is also possible to analyze according to the steric parameter of thing
Go out the number change of damper, it is necessary to when designing certain attribute of detection damper, I designs coherent detection according to the actual requirements
Model, these cannot not design a model fixedly and select particular model, are defined by that can calculate fault message using steric parameter, this
The central scope of invention aims to provide a kind of method and obtains damper steric parameter, and the design of fault message is analyzed from steric parameter
Scheme, above-mentioned steric information is to refer to the three-dimensional coordinate information obtained in such scheme.
Embodiment 2
On the basis of above-described embodiment, it is preferred that shift failure, damage failure because damper is usually present, come off
Failure, foreign matter interference failure, the damper status attribute that we design is related above-mentioned failure, is specifically expressed as follows:It is described
Damper status attribute set S includes ι, m, r, n;ι represents that damper (can represent shifting with fixed reference according to the distance between thing
Position failure);M represents an iron hammer quantity for damper, and under normal circumstances, damper has two iron hammers (can to represent damage
Bad failure);R represents chaff interference to the distance (can represent that foreign matter disturbs failure) of damper;N represents the anti-of a certain section of overhead route
Vibration hammer number (can represent release failure);Then initial value S0Including ι0、m0、r0、n0;Currency StIncluding ιt、mt、rt、nt;Institute
Stating changing value △ S includes △ ι, △ m, △ r, △ n, wherein, △ ι=ιt- ι0, △ m=mt- m0, △ r=rt- r0, △ n=
nt- n0。ι0Damper shines the distance between thing, m with fixed reference under representing original state0Represent the next vibrationproof of original state
The iron hammer quantity of hammer, r0Minimum standard value, the n of the distance of chaff interference and damper under expression original state0Represent initial shape
The damper number of overhead route under state;ιtDamper shines the distance between thing, m with fixed reference under representing current statetRepresent and work as
Iron hammer quantity, the r of the preceding next damper of statetDistance, the n of chaff interference and damper under expression current statetRepresent current shape
The damper number of overhead route under state.
Wherein, because damper displacement failure is difficult to judge, therefore, one fixed reference of present invention design shines thing, with admittedly
The fixed change with reference to according to the distance between thing and damper is recognized between damper with the presence or absence of displacement, in order to know
Do not go out the shift amount of damper, our preferred fixed references are overhead route wire clamp according to thing, due to overhead route wire clamp and damper all
It is arranged on same thread guide path, therefore is advantageously provided with and identifies shift amount, it is also possible to fixed reference is set to according to thing
Any damper or one in addition the reference substance of setting, other objects that can also be on tower bar, can also be tower bar in itself.Together
When, damper has two iron hammers under normal circumstances, and in the case of thunderbolt or other, iron hammer occurs that damage is dropped so that vibrationproof
Hammer is uneven, therefore, it can obtain depth map by the three-dimensional coordinate of damper, from depth map analysis iron hammer quantity, by preceding
Comparative analysis afterwards has and damages failure, therefore, damaging failure can be represented by detecting iron hammer number change.Further, since anti-
Vibration hammer drop failure occur when, the quantity of damper can change, therefore the number of damper can be judged by 3 D stereo parameter
Amount, represents that damper drops failure by the number change of damper.Due to requiring with damper center with r0In radius
Chaff interference must not occur, therefore, when chaff interference has been detected, we are not chaff interferences again by 3 D stereo parameter
Steric information, chaff interference calculate between damper by the steric information of chaff interference and the steric information of damper
Distance, so as to detect chaff interference failure, wherein r0Belong in advance according to the value set about requirement, by rtWith r0Difference
Value represents chaff interference failure.
Preferably, following process is included into Binocular Stereo Vision System using two image mechanisms:
Build binocular stereo vision hardware system:Two video cameras are fixed on optical table with the spacing of parallax range B
On;
Shoot scaling board image sets:Make longitude and latitude case marker fixed board complete imaging in two cameras, shoot multigroup different attitudes
Under longitude and latitude case marker fixed board image;
Obtain parameter:Coordinate system with any camera extracts longitude and latitude case marker fixed board image angle point, with most as world coordinate system
Small square law obtains the distortion parameter of the inner parameter matrix, external parameter matrix and left and right camera lens of left and right camera.
Preferably, the abnormal failure detection of damper includes following process:
Obtain background image:Turn off video camera A, the image obtained by video camera B is used as background image;
Abnormality detection:Current first two field picture is shot by video camera B, abnormality detection is carried out using Background difference, there is different
Often turn fault detect during detection;
Fault detect:Current second two field picture is shot by video camera B, fault detect is carried out using Background difference.
Carrying out abnormality detection using Background difference includes following process:
Current first two field picture I that video camera B is shott(x, y) and background image ut(x, y) does subtraction and obtains
Background difference;Background difference is Dt(x, y)=| It(x,y)-ut(x,y)|;Work as DtWhen (x, y) >=ε, ε are unusual determination threshold value,
Then it is judged to exception occur;
Carrying out fault detect using Background difference includes following process:
Obtain current second two field picture I of video camera B shootingst+1(x, y) and background image ut(x, y) does subtraction and obtains
To fault detect background difference;Fault detect background difference is Dt+1(x, y)=| It+1(x, y)-ut(x, y) |;Work as Dt+1(x,y)
>=ε, ε are unusual determination threshold value, then be judged to break down.
By it is above-mentioned as can be seen that the present invention in the abnormal failure detection process of damper, employ abnormality detection and
Two processes of fault detect, its reason is, because detection environment is outside exploitation formula environment, can usually there is bird, leaf etc. miscellaneous
Thing is by camera visual range, but the appearance process of these debris is very of short duration, therefore, if only carrying out exception successively
Detection, then can trigger many wrong reports, and the operation of current failure analysis is just opened under the only 2 times regulations of detection of the present invention, keep away
Exempt from system and be in frequent starting state, whole system is standby under can maintaining low power consumpting state before the fault detect, also will not
There is multiple wrong report, reduce line walking number of times, second two field picture in the present invention refers to that first two field picture is later than on the time
Image, and necessarily refer not to the adjacent two field picture after first two field picture.
Preferably, it is described below down, the present invention obtains the process of three-dimensional coordinate:
Obtain damper three-dimensional coordinate, fixed reference by initial image information includes following process according to thing three-dimensional coordinate:
In an initial condition, left image, right image are obtained by left video camera, right video camera respectively, forms initial pictures
Information;
Left image, right image are carried out to eliminate distortion and polar curve correction process, left image and right figure after making elimination distort
As strictly row is corresponding;
Image preprocessing is carried out to left image and right image;
Left image and right image characteristic point are extracted respectively;
Matching left image and the characteristic point of right image, obtain multigroup matching double points;
Obtain image three-dimensional coordinate:Left image is obtained with right image in respective pixel using binocular stereo vision measurement model
The three-dimensional coordinate of coordinate system;
Realize that pixel coordinate is tied to the conversion of world coordinate system:The characteristic point of left image is set in the pixel coordinate system of oneself
Pixel coordinate (xl, yl), the characteristic point of right image the pixel coordinate system of oneself pixel coordinate (xr, yr), (xl, yl) with
(xr, yr) matching double points are formed, in the space orientation of image, image measurement is distant, a minor variations of pixel coordinate
Also huge measurement error can be caused.Present invention selection is extracted subpixel coordinates and forms matching again after the completion of image sequence Point matching
Point is right so that the three-dimensional measuring result of image is more accurate, image scene object more accurate positioning.
Three dimensional space coordinate (X of the matching double points in world coordinate systemW,YW,ZW) relation be shown below:
Wherein, xlAnd xrRepresent that left image, right image obtain horizontal seat of the matching double points under pixel coordinate system after matching
Mark, ylAnd yrRepresent that left image, right image obtain ordinate of the matching double points under pixel coordinate system after matching, B represents that a left side is taken the photograph
Parallax range between camera and right video camera, f represents the focal length of video camera where world coordinate system, and B and f are obtained according to camera calibration
Arrive;
If matching double points are the matching double points of damper, three dimensional space coordinate of the damper in world coordinate system
(X0(stockbridge), Y0(stockbridge), Z0(stockbridge)) relation be shown below:
If matching double points be fixed reference according to thing matching double points, fixed reference according to thing world coordinate system three-dimensional space
Between coordinate (X0(clamp), Y0(clamp), Z0(clamp)) relation be shown below:
Obtain the initial value S of damper status attribute set S0Including following process:
Obtain l0:Detect that damper and fixed reference shine thing, according to damper, fixed reference according to thing in world coordinate system
Three dimensional space coordinate calculate damper and fixed reference according to the distance between thing l0, wherein,
Obtain m0:Directly configure m0=2 obtain depth map and obtain 3D point cloud data again by the three-dimensional coordinate of damper
The iron hammer number m of damper is detected by 3D matching process0, wherein m0Represent an iron hammer quantity for damper;
Obtain r0:Directly configure r0Value, r0Represent chaff interference to the distance of damper, r now0It is damper center
Radius;
Obtain n0:The electric wire in image is detected, window is then found with a damper and is slided along electric wire, often detected
To a damper, damper quantity adds 1, final n0Represent a certain section of damper number of overhead route;
ι0、m0、r0、n0It is the initial value S of damper status attribute set S0。
Following process is included according to thing three-dimensional coordinate by present image information acquisition damper three-dimensional coordinate, fixed reference:
Under current state, left image, right image are obtained by left video camera, right video camera respectively, form present image
Information;
Left image, right image are carried out to eliminate distortion and polar curve correction process, left image and right figure after making elimination distort
As strictly row is corresponding;
Image preprocessing is carried out to left image and right image;
Left image and right image characteristic point are extracted respectively;
Matching left image and the characteristic point of right image, obtain multigroup matching double points;
Obtain image three-dimensional coordinate:Left image is obtained with right image in respective pixel using binocular stereo vision measurement model
The three-dimensional coordinate of coordinate system;
Realize that pixel coordinate is tied to the conversion of world coordinate system:The characteristic point of left image is set in the pixel coordinate system of oneself
Pixel coordinate (xl, yl), the characteristic point of right image the pixel coordinate system of oneself pixel coordinate (xr, yr), (xl, yl) with
(xr, yr) form matching double points, three dimensional space coordinate (X of the matching double points in world coordinate systemW,YW,ZW) relation such as following formula institute
Show:
Wherein, xlAnd xrRepresent that left image, right image obtain horizontal seat of the matching double points under pixel coordinate system after matching
Mark, ylAnd yrRepresent that left image, right image obtain ordinate of the matching double points under pixel coordinate system after matching, B represents that a left side is taken the photograph
Parallax range between camera and right video camera, f represents the focal length of video camera where world coordinate system, and B and f are obtained according to camera calibration
Arrive;
If matching double points are the matching double points of damper, three dimensional space coordinate of the damper in world coordinate system
(Xt(stockbridge), Yt(stockbridge), Zt(stockbridge)) relation be shown below:
If matching double points be fixed reference according to thing matching double points, fixed reference according to thing world coordinate system three-dimensional space
Between coordinate (Xt(clamp), Yt(clamp), Zt(clamp)) relation be shown below:
Whether identification present image information has damper chaff interference, if with the presence of damper chaff interference, also including obtaining
Take damper chaff interference three-dimensional coordinate process as follows:
The matching double points that the matching double points are damper chaff interference are recognized, then damper chaff interference is in world coordinate system
Three dimensional space coordinate (Xt(distractors), Yt(distractors), Zt(distractors)) relation be shown below:
Obtain the currency S of damper status attribute set StIncluding following process:
Obtain lt:Detect that damper and fixed reference shine thing, according to damper, fixed reference according to thing in world coordinate system
Three dimensional space coordinate calculate damper and fixed reference according to the distance between thing lt, wherein
Obtain mt:3D point cloud data are obtained by the three-dimensional coordinate acquisition depth map of damper to be detected by 3D matching process again
Go out the iron hammer number m of dampert, wherein mtRepresent an iron hammer quantity for damper;
Obtain rt:
If centered on damper, with r0Have to be identified present image information using frame difference method in the range of radius
Damper chaff interference, then calculate damper chaff interference to damper apart from rt, r0It is the value of directly configuration, wherein,
If centered on damper, with r0Do not have to be identified present image information using frame difference method in the range of radius
Damper chaff interference, then assign damper chaff interference to damper apart from rt, assign rt=r0;
Obtain nt:The electric wire in image is detected, window is then found with a damper and is slided along electric wire, often detected
To a damper, damper quantity adds 1, final ntRepresent a certain section of damper number of overhead route;
lt、mt、rt、ntIt is the currency S of damper status attribute set St。
Judge that the situation for obtaining damper failure includes procedure below according to △ S:
Output displacement failure:If Δ l ∈ [a, b], then illustrate that damper, in the absence of displacement failure, otherwise, has displacement event
Barrier, wherein, a, b is respectively the lower and upper limit of damper and fixed reference according to the safe distance scope between thing, according to △ ι's
Value represents damper translocation distance;
Output iron hammer damages failure:If Δ m=0, show damper in the absence of damaged condition, otherwise, damper is deposited
Failure is being damaged, the value according to △ m represents that iron hammer damages quantity;
Output damper chaff interference failure:If Δ r≤0, illustrate that damper has foreign matter interference failure, according to △ r
Value represent chaff interference to the distance of damper, otherwise in the absence of damper chaff interference failure;
Output damper release failure:If Δ n ≠ 0, illustrate there is damper release failure, the value table according to △ n
Show the quantity that damper comes off, otherwise in the absence of damper release failure.
The characteristics of property according to damper each fault category, can be judged obtain the feelings of failure by state changing features amount
Condition.It is specifically described as:
1) the distance between damper and overhead route wire clamp have certain safety standard, if this distance is not in safety
Within standard, then it is assumed that be to be shifted over, that is, there is displacement failure.I.e.:If Δ l ∈ [a, b], then illustrate that damper does not exist
, otherwise, there is displacement failure in displacement failure.Wherein, a, b are respectively the safe distance scope between damper and overhead route wire clamp
Lower and upper limit.
2) damper is made up of a bit of metal of a geometrical clamp and two with two iron hammers, and two must have two iron
Hammer, if only one of which or do not had, referred to as damages.Therefore the RM for damaging failure is:
If Δ m=0, show damper in the absence of damaged condition, otherwise, damper has damage failure.
3) in the certain limit around damper, do not allow the presence of other object, otherwise damper is interfered,
I.e. there is foreign matter interference failure in damper.Damper foreign matter interference Fault Identification mode be:If Δ r≤0, illustrates damper
There is foreign matter interference failure, do not exist otherwise.
4) need damper quantity to be mounted more than one in some overhead routes, if the quantity for detecting is not standard
Damper quantity, then must be that some dampers have completely fallen off.Damper release failure decision method is:
If Δ n ≠ 0, illustrate there is damper release failure, do not exist otherwise.
Damper fault type list is finally exported, and failure concrete condition is exported.For example, a=0.01 meters, b=
0.05 meter, if damper malfunction property set is Δ S={ 0.03, -1, -0.5,0 }, then corresponding output result is:It is anti-
There is damage failure and foreign matter interference failure in vibration hammer.If Δ S={ 3,0,0,0 }, then illustrate that damper has displacement failure, and
And shift value is 3 meters.
Claims (10)
1. a kind of damper On-line Fault Detection recognition methods based on binocular vision image, it is characterised in that including following step
Suddenly:
Using two image mechanisms into Binocular Stereo Vision System;
Set up a damper status attribute set S;
In an initial condition, being obtained using two video cameras has damper, fixed reference according to the initial image information of thing, passes through
Acquisition damper three-dimensional coordinate, the fixed reference of initial image information are according to thing three-dimensional coordinate so as to obtain damper status attribute collection
Close the initial value S of S0;
Under current state, the abnormal failure detection of damper, it is found that when abnormal failure occurs in damper, using two video cameras
Obtain with damper, fixed reference according to thing present image information, by present image information acquisition damper three-dimensional coordinate,
Fixed reference is according to thing three-dimensional coordinate so as to obtain the currency S of damper status attribute set St;
Damper fault type recognition:According to currency StWith initial value S0Calculate the changing value △ S of damper status attribute, root
Judge to obtain the situation of damper failure according to △ S.
2. a kind of damper On-line Fault Detection recognition methods based on binocular vision image according to claim 1, its
It is characterised by, the damper status attribute set S includes ι, m, r, n;ι represent damper and fixed reference according between thing away from
From;M represents an iron hammer quantity for damper, and under normal circumstances, a damper has two iron hammers;R represents chaff interference to anti-
The distance of vibration hammer;N represents a certain section of damper number of overhead route;Then initial value S0Including ι0、m0、r0、n0;Currency StIncluding
ιt、mt、rt、nt;The changing value △ S include △ ι, △ m, △ r, △ n, wherein, △ ι=ιt- ι0, △ m=mt- m0, △ r=
rt- r0, △ n=nt- n0。
3. a kind of damper On-line Fault Detection recognition methods based on binocular vision image according to claim 1, its
It is characterised by, following process is included into Binocular Stereo Vision System using two image mechanisms:
Build binocular stereo vision hardware system:Two video cameras are fixed on optical table with the spacing of parallax range B;
Shoot scaling board image sets:Make longitude and latitude case marker fixed board complete imaging in two cameras, shoot under multigroup different attitudes
The image of longitude and latitude case marker fixed board;
Obtain parameter:Coordinate system with any camera extracts longitude and latitude case marker fixed board image angle point, with a most young waiter in a wineshop or an inn as world coordinate system
Multiplication obtains the distortion parameter of the inner parameter matrix, external parameter matrix and left and right camera lens of left and right camera.
4. a kind of damper On-line Fault Detection recognition methods based on binocular vision image according to claim 1, its
It is characterised by, the abnormal failure detection of damper includes following process:
Obtain background image:Turn off video camera A, the image obtained by video camera B is used as background image;
Abnormality detection:Current first two field picture is shot by video camera B, abnormality detection is carried out using Background difference, there is abnormal inspection
Turn fault detect during survey;
Fault detect:Current second two field picture is shot by video camera B, fault detect is carried out using Background difference;Background image
Time of the time earlier than current first two field picture, the time of current first two field picture is earlier than current second two field picture
Time.
5. a kind of damper On-line Fault Detection recognition methods based on binocular vision image according to claim 4, its
It is characterised by,
Carrying out abnormality detection using Background difference includes following process:
Current first two field picture I that video camera B is shott(x, y) and background image ut(x, y) does subtraction and obtains background
Difference;Background difference is Dt(x, y)=| It(x,y)-ut(x,y)|;Work as DtDuring (x, y) >=ε, then it is judged to exception occur, ε is
Unusual determination threshold value;
Carrying out fault detect using Background difference includes following process:
Obtain current second two field picture I of video camera B shootingst+1(x, y) and background image ut(x, y) does subtraction and obtains failure
Detection background difference;Fault detect background difference is Dt+1(x, y)=| It+1(x, y)-ut(x, y) |;Work as Dt+1During (x, y) >=ε,
Then it is judged to break down, ε is unusual determination threshold value.
6. a kind of damper On-line Fault Detection based on binocular vision image according to any one in claim 1-5
Recognition methods, it is characterised in that damper three-dimensional coordinate, fixed reference are obtained by initial image information and shines thing three-dimensional coordinate bag
Include following process:
In an initial condition, left image, right image are obtained by left video camera, right video camera respectively, forms initial image information;
Left image, right image are carried out to eliminate distortion and polar curve correction process, the left image after making elimination distort is tight with right image
Lattice ground row correspondence;
Image preprocessing is carried out to left image and right image;
Left image and right image characteristic point are extracted respectively;
Matching left image and the characteristic point of right image, obtain multigroup matching double points;
Obtain image three-dimensional coordinate:Left image is obtained with right image in respective pixel coordinate using binocular stereo vision measurement model
The three-dimensional coordinate of system;
Realize that pixel coordinate is tied to the conversion of world coordinate system:The picture of the characteristic point in the pixel coordinate system of oneself of left image is set
Plain coordinate (xl, yl), the characteristic point of right image the pixel coordinate system of oneself pixel coordinate (xr, yr), (xl, yl) and (xr, yr)
Form matching double points, three dimensional space coordinate (X of the matching double points in world coordinate systemW,YW,ZW) relation be shown below:
Wherein, xlAnd xrRepresent that left image, right image obtain abscissa of the matching double points under pixel coordinate system, y after matchingl
And yrRepresent that left image, right image obtain ordinate of the matching double points under pixel coordinate system after matching, B represents left video camera
With the parallax range between right video camera, f represents the focal length of video camera where world coordinate system, and B and f are obtained according to camera calibration;
If matching double points are the matching double points of damper, three dimensional space coordinate of the damper in world coordinate system
(X0(stockbridge), Y0(stockbridge), Z0(stockbridge)) relation be shown below:
If matching double points are matching double points of the fixed reference according to thing, three dimensions of the fixed reference according to thing in world coordinate system is sat
Mark (X0(clamp), Y0(clamp), Z0(clamp)) relation be shown below:
7. a kind of damper On-line Fault Detection recognition methods based on binocular vision image according to claim 6, its
It is characterised by, obtains the initial value S of damper status attribute set S0Including following process:
Obtain l0:Detect that damper and fixed reference shine thing, according to damper, fixed reference according to thing the three of world coordinate system
Dimension space coordinate calculates damper with fixed reference according to the distance between thing l0, wherein,
Obtain m0:Directly configure m0=2 obtain depth map and obtain 3D point cloud data and pass through again by the three-dimensional coordinate of damper
3D matching process detects the iron hammer number m of damper0, wherein m0Represent an iron hammer quantity for damper;
Obtain r0:Directly configure r0Value, r0Represent chaff interference to the distance of damper, r now0It is the radius at damper center;
Obtain n0:The electric wire in image is detected, window is then found with a damper and is slided along electric wire, often detect one
Individual damper, damper quantity adds 1, final n0Represent a certain section of damper number of overhead route;
ι0、m0、r0、n0It is the initial value S of damper status attribute set S0。
8. a kind of damper On-line Fault Detection based on binocular vision image according to any one in claim 1-5
Recognition methods, it is characterised in that thing three-dimensional coordinate bag is shone by present image information acquisition damper three-dimensional coordinate, fixed reference
Include following process:
Under current state, left image, right image are obtained by left video camera, right video camera respectively, form present image information;
Left image, right image are carried out to eliminate distortion and polar curve correction process, the left image after making elimination distort is tight with right image
Lattice ground row correspondence;
Image preprocessing is carried out to left image and right image;
Left image and right image characteristic point are extracted respectively;
Matching left image and the characteristic point of right image, obtain multigroup matching double points;
Obtain image three-dimensional coordinate:Left image is obtained with right image in respective pixel coordinate using binocular stereo vision measurement model
The three-dimensional coordinate of system;
Realize that pixel coordinate is tied to the conversion of world coordinate system:The picture of the characteristic point in the pixel coordinate system of oneself of left image is set
Plain coordinate (xl, yl), the characteristic point of right image the pixel coordinate system of oneself pixel coordinate (xr, yr), (xl, yl) and (xr, yr)
Form matching double points, three dimensional space coordinate (X of the matching double points in world coordinate systemW,YW,ZW) relation be shown below:
Wherein, xlAnd xrRepresent that left image, right image obtain abscissa of the matching double points under pixel coordinate system, y after matchingl
And yrRepresent that left image, right image obtain ordinate of the matching double points under pixel coordinate system after matching, B represents left video camera
With the parallax range between right video camera, f represents the focal length of video camera where world coordinate system, and B and f are obtained according to camera calibration;
If matching double points are the matching double points of damper, three dimensional space coordinate of the damper in world coordinate system
(Xt(stockbridge), Yt(stockbridge), Zt(stockbridge)) relation be shown below:
If matching double points are matching double points of the fixed reference according to thing, three dimensions of the fixed reference according to thing in world coordinate system is sat
Mark (Xt(clamp), Yt(clamp), Zt(clamp)) relation be shown below:
Whether identification present image information has damper chaff interference, if with the presence of damper chaff interference, also including obtaining anti-
Vibration hammer chaff interference three-dimensional coordinate process is as follows:
The matching double points that the matching double points are damper chaff interference are recognized, then three-dimensional of the damper chaff interference in world coordinate system
Space coordinates (Xt(distractors), Yt(distractors), Zt(distractors)) relation be shown below:
9. a kind of damper On-line Fault Detection recognition methods based on binocular vision image according to claim 8, its
It is characterised by, obtains the currency S of damper status attribute set StIncluding following process:
Obtain lt:Detect that damper and fixed reference shine thing, according to damper, fixed reference according to thing the three of world coordinate system
Dimension space coordinate calculates damper with fixed reference according to the distance between thing lt, wherein
Obtain mt:By damper three-dimensional coordinate obtain depth map obtain 3D point cloud data detected by 3D matching process again it is anti-
The iron hammer number m of vibration hammert, wherein mtRepresent an iron hammer quantity for damper;
Obtain rt:
If centered on damper, with r0There is vibrationproof to be identified present image information using frame difference method in the range of radius
Hammer chaff interference, then calculate damper chaff interference to damper apart from rt, r0It is the value of directly configuration, wherein,
If centered on damper, with r0There is no vibrationproof to be identified present image information using frame difference method in the range of radius
Hammer chaff interference, then assign damper chaff interference to damper apart from rt, assign rt=r0;
Obtain nt:The electric wire in image is detected, window is then found with a damper and is slided along electric wire, often detect one
Individual damper, damper quantity adds 1, final ntRepresent a certain section of damper number of overhead route;
lt、mt、rt、ntIt is the currency S of damper status attribute set St。
10. a kind of damper On-line Fault Detection recognition methods based on binocular vision image according to claim 2, its
It is characterised by, judges that the situation for obtaining damper failure includes procedure below according to △ S:
Output displacement failure:If Δ l ∈ [a, b], then illustrate that damper, in the absence of displacement failure, otherwise, has displacement failure, its
In, a, b is respectively the lower and upper limit of damper and fixed reference according to the safe distance scope between thing, the value table according to △ ι
Show damper translocation distance;
Output iron hammer damages failure:If Δ m=0, show damper in the absence of damaged condition, otherwise, damper has damage
Bad failure, the value according to △ m represents that iron hammer damages quantity;
Output damper chaff interference failure:If Δ r≤0, illustrate that damper has foreign matter interference failure, according to the value of △ r
Represent chaff interference to the distance of damper;Do not exist chaff interference failure otherwise;
Output damper release failure:If Δ n ≠ 0, illustrate there is damper release failure, the value according to △ n represents anti-
The quantity that vibration hammer comes off, otherwise in the absence of damper release failure.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611116535.0A CN106709905B (en) | 2016-12-07 | 2016-12-07 | Vibration damper fault online detection and identification method based on binocular vision image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611116535.0A CN106709905B (en) | 2016-12-07 | 2016-12-07 | Vibration damper fault online detection and identification method based on binocular vision image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106709905A true CN106709905A (en) | 2017-05-24 |
CN106709905B CN106709905B (en) | 2020-03-31 |
Family
ID=58936228
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611116535.0A Active CN106709905B (en) | 2016-12-07 | 2016-12-07 | Vibration damper fault online detection and identification method based on binocular vision image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106709905B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107992820A (en) * | 2017-11-29 | 2018-05-04 | 北京伟景智能科技有限公司 | Counter automatic selling method based on binocular vision |
CN109165318A (en) * | 2018-08-13 | 2019-01-08 | 洛阳视距智能科技有限公司 | A kind of damper data set construction method towards intelligent patrol detection |
CN110956658A (en) * | 2019-10-26 | 2020-04-03 | 湖北讯獒信息工程有限公司 | Image feature recognition algorithm acceleration method in binocular stereoscopic vision system |
CN111738907A (en) * | 2020-06-08 | 2020-10-02 | 广州运达智能科技有限公司 | Train pantograph detection method based on binocular calibration and image algorithm |
CN112014688A (en) * | 2020-09-01 | 2020-12-01 | 云南电网有限责任公司电力科学研究院 | Method and device for detecting slippage fault of damper |
CN114170131A (en) * | 2021-10-20 | 2022-03-11 | 国网福建省电力有限公司信息通信分公司 | Transmission line equipment defect detection method and system |
CN114445335A (en) * | 2021-12-22 | 2022-05-06 | 武汉易思达科技有限公司 | Vehicle running state monitoring method and system based on binocular machine vision |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102621419A (en) * | 2012-03-28 | 2012-08-01 | 山东省电力学校 | Method for automatically recognizing and monitoring line electrical equipment based on laser and binocular vision image |
CN103407407A (en) * | 2013-08-28 | 2013-11-27 | 沈阳工业大学 | Automobile safety distance warning device and method based on binocular stereo vision |
CN103559703A (en) * | 2013-10-08 | 2014-02-05 | 中南大学 | Crane barrier monitoring and prewarning method and system based on binocular vision |
US20150022642A1 (en) * | 2013-07-16 | 2015-01-22 | Texas Instruments Incorporated | Super-Resolution in Structured Light Imaging |
CN105718964A (en) * | 2014-12-02 | 2016-06-29 | 中国科学院沈阳自动化研究所 | Transmission line vibration damper visual detection method |
-
2016
- 2016-12-07 CN CN201611116535.0A patent/CN106709905B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102621419A (en) * | 2012-03-28 | 2012-08-01 | 山东省电力学校 | Method for automatically recognizing and monitoring line electrical equipment based on laser and binocular vision image |
US20150022642A1 (en) * | 2013-07-16 | 2015-01-22 | Texas Instruments Incorporated | Super-Resolution in Structured Light Imaging |
CN103407407A (en) * | 2013-08-28 | 2013-11-27 | 沈阳工业大学 | Automobile safety distance warning device and method based on binocular stereo vision |
CN103559703A (en) * | 2013-10-08 | 2014-02-05 | 中南大学 | Crane barrier monitoring and prewarning method and system based on binocular vision |
CN105718964A (en) * | 2014-12-02 | 2016-06-29 | 中国科学院沈阳自动化研究所 | Transmission line vibration damper visual detection method |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107992820A (en) * | 2017-11-29 | 2018-05-04 | 北京伟景智能科技有限公司 | Counter automatic selling method based on binocular vision |
CN109165318A (en) * | 2018-08-13 | 2019-01-08 | 洛阳视距智能科技有限公司 | A kind of damper data set construction method towards intelligent patrol detection |
CN110956658A (en) * | 2019-10-26 | 2020-04-03 | 湖北讯獒信息工程有限公司 | Image feature recognition algorithm acceleration method in binocular stereoscopic vision system |
CN111738907A (en) * | 2020-06-08 | 2020-10-02 | 广州运达智能科技有限公司 | Train pantograph detection method based on binocular calibration and image algorithm |
CN111738907B (en) * | 2020-06-08 | 2022-08-23 | 广州运达智能科技有限公司 | Train pantograph detection method based on binocular calibration and image algorithm |
CN112014688A (en) * | 2020-09-01 | 2020-12-01 | 云南电网有限责任公司电力科学研究院 | Method and device for detecting slippage fault of damper |
CN112014688B (en) * | 2020-09-01 | 2022-09-02 | 云南电网有限责任公司电力科学研究院 | Method and device for detecting slippage fault of damper |
CN114170131A (en) * | 2021-10-20 | 2022-03-11 | 国网福建省电力有限公司信息通信分公司 | Transmission line equipment defect detection method and system |
CN114445335A (en) * | 2021-12-22 | 2022-05-06 | 武汉易思达科技有限公司 | Vehicle running state monitoring method and system based on binocular machine vision |
CN114445335B (en) * | 2021-12-22 | 2024-04-12 | 武汉易思达科技有限公司 | Vehicle running state monitoring method based on binocular machine vision |
Also Published As
Publication number | Publication date |
---|---|
CN106709905B (en) | 2020-03-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106709905A (en) | Vibration-proof hammer fault online detection and identification method based on binocular vision image | |
CN111339882B (en) | Power transmission line hidden danger detection method based on example segmentation | |
CN110533722A (en) | A kind of the robot fast relocation method and system of view-based access control model dictionary | |
CN103716579B (en) | Video monitoring method and system | |
Yang et al. | Deep learning‐based bolt loosening detection for wind turbine towers | |
CN106372662A (en) | Helmet wearing detection method and device, camera, and server | |
CN106356757A (en) | Method for inspecting electric power lines by aid of unmanned aerial vehicle on basis of human vision characteristics | |
CN107564062A (en) | Pose method for detecting abnormality and device | |
CN109829908B (en) | Binocular image-based method and device for detecting safety distance of ground object below power line | |
CN111160125A (en) | Railway foreign matter intrusion detection method based on railway monitoring | |
CN104007240A (en) | Fusion positioning technology based on binocular recognition and electronic nose network gas detection | |
CN104316782B (en) | pantograph monitoring method and system | |
CN111222478A (en) | Construction site safety protection detection method and system | |
CN106649598A (en) | Detection method for icing thickness of power transmission line | |
CN102298816B (en) | Fire early warning method for marine engine room based on multi-source fusion | |
CN103973697B (en) | A kind of thing network sensing layer intrusion detection method | |
CN108596108A (en) | Method for detecting change of remote sensing image of taking photo by plane based on the study of triple semantic relation | |
CN114240868A (en) | Unmanned aerial vehicle-based inspection analysis system and method | |
CN102621154A (en) | Method and device for automatically detecting cloth defects on line based on improved differential box multi-fractal algorithm | |
CN102231844A (en) | Video image fusion performance evaluation method based on structure similarity and human vision | |
CN117010532A (en) | Comprehensive pipe gallery fire trend prediction method based on multi-mode deep learning | |
CN115547003A (en) | External damage prevention alarm method and system for power transmission line | |
CN112330675B (en) | Traffic road image atmospheric visibility detection method based on AOD-Net | |
CN112509033B (en) | Automatic iron tower inclination detection method and device based on image processing | |
CN111753587A (en) | Method and device for detecting falling to ground |
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 |