CN109029539A - A kind of municipal rail train pantograph goat's horn monitoring device and method - Google Patents
A kind of municipal rail train pantograph goat's horn monitoring device and method Download PDFInfo
- Publication number
- CN109029539A CN109029539A CN201810424208.4A CN201810424208A CN109029539A CN 109029539 A CN109029539 A CN 109029539A CN 201810424208 A CN201810424208 A CN 201810424208A CN 109029539 A CN109029539 A CN 109029539A
- Authority
- CN
- China
- Prior art keywords
- horn
- goat
- image
- pantograph
- template
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- 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
- G06T2207/10004—Still image; Photographic image
Abstract
The invention discloses a kind of municipal rail train pantograph goat's horn monitoring device and methods.The device includes field control unit, field data acquisition unit, remote transmission unit, data analysis unit, data storage and release unit and client access unit.Method are as follows: pantograph original image is obtained by industrial camera first；Secondly original image is filtered using Gaussian filter algorithm；Pantograph pan or so extreme coordinates are obtained further according to endpoint template matching, coarse positioning further is carried out to the goat's horn in the image after filtering processing；Then, according to the goat's horn image after coarse positioning, translation rotation transformation is carried out to the goat's horn template of engineer, it is matched utmostly with the goat's horn after coarse positioning, realizes the fine positioning of goat's horn；Finally differentiated using the goat's horn basis of characterization based on edge direction to whether the goat's horn after fine positioning lacks.The present invention can be realized the contactless monitoring to goat's horn, have the advantages that high-efficient, precision is high, highly-safe.
Description
Technical field
The present invention relates to traffic safety engineering monitoring technical field, especially a kind of municipal rail train pantograph goat's horn monitoring dress
It sets and method.
Background technique
In recent years, municipal rail train relied on its quick, comfortable, energy conservation, efficient advantage, had obtained hair at full speed in China
Exhibition.Meanwhile the safety problem of train on-line operation is also increasingly significant.Municipal rail train pantograph be by electric energy from contact transfers on network
Electric device to municipal rail train to drive it to operate normally.The quality of pantograph pan directly affects the safety fortune of municipal rail train
Row, the missing of pantograph goat's horn, it will leading to train, ratio of contact loss increases in the process of running, influences the normal power supply of municipal rail train
Safety accident is caused, resulting arc discharge can also be further exacerbated by the abrasion of pantograph pan and contact line.With state
The attention of municipal rail train construction is continuously improved in family, and to the reliability of pantograph, safety, more stringent requirements are proposed, therefore right
The monitoring of pantograph goat's horn is of great significance.
The monitoring method of country's pantograph state mainly includes personal monitoring's method at present.Personal monitoring needs municipal rail train to return
Library parking, contact net power-off not only reduce the operational efficiency of municipal rail train, and staff needs to climb up roof monitoring,
Heavy workload, there are some potential safety problemss.Domestic colleges and universities and each company have also started the research monitored to pantograph, mainly have
Image monitoring method, laser monitoring method etc., research achievement is mostly under test, shadow of the monitoring result vulnerable to external environment
It rings, and is monitored to pantograph pan, it is then relatively fewer to the study on monitoring of pantograph goat's horn.
Summary of the invention
The purpose of the present invention is to provide it is a kind of be simple and efficient, municipal rail train pantograph goat's horn monitoring device at low cost and
Method provides technical support for the monitoring and maintenance of pantograph goat's horn.
The technical solution for realizing the aim of the invention is as follows: a kind of municipal rail train pantograph goat's horn monitoring device, including existing
Field control unit, field data acquisition unit, remote transmission unit, data analysis unit, data storage and release unit, client
Hold access unit, in which:
Field control unit detects the arrival of train and obtains train vehicle number information；
Field data acquisition unit, the acquisition for pantograph original image；
Remote transmission unit, on site with the data in equipment room transmit；
Data analysis unit handles the image received；
Data storage and release unit, storage and publication for data；
Client access unit, for remotely accessing.
Further, the field control unit includes PLC, wheel axial sensor D, number identification device AEI,
In:
The wheel axial sensor D is mounted on the inside of track forefront, for detecting arrival and the generation of municipal rail train
Corresponding signal；
PLC is used to receive the signal of wheel axial sensor D transmission to control number identification device AEI；
Number identification device AEI obtains the license number of municipal rail train by Train number recognition antenna.
Further, the field data acquisition unit includes the first industry flash of light lamp group L1-1, the second industry flash of light
Lamp group L1-2, third industry flash of light lamp group L2-1, the 4th industry flash of light lamp group L2-2, the first photoelectric sensor P1-1, the second light
Electric transducer P1-2, third photoelectric sensor P2-1, the 4th photoelectric sensor P2-2, the first industrial camera C1, the second industrial phase
Machine C2；
The first photoelectric sensor P1-1 and the second photoelectric sensor P1-2 is symmetrically arranged on the top of two siding tracks, the
Three photoelectric sensor P2-1 and the 4th photoelectric sensor P2-2 are symmetrically arranged on the top of two siding tracks, and each photoelectric sensor is high
Degree is consistent with height where slide plate when pantograph rising bow, be used for external trigger the first~bis- industrial camera C1, C2 and first~
Four industry flash of light lamp group L1-1, L1-2, L2-1, L2-2；The first industry industry flash of light of flash of light lamp group L1-1 and second lamp group L1-2
It is respectively arranged in the top of two siding tracks, the third industry industry flash of light of flash of light lamp group L2-1 and the 4th lamp group L2-2 is respectively arranged in
The top of two siding tracks, industry flash of light lamp group provide light filling for the shooting of the first industrial camera C1, the second industrial camera C2；First
Industrial camera C1, the second industrial camera C2 are center line camera, are installed on right above track centerline, for acquiring image.
Further, the collecting flowchart of the field control unit and field data acquisition unit are as follows: when wheel axle position
When sensor D detects that train arrives, number identification device AEI is opened；When train logo passes through number identification device AEI,
Number identification device AEI reads the vehicle number information of the train, and is transmitted to data analysis unit by remote transmission unit；When
When one photoelectric sensor P1-1 and the second photoelectric sensor P1-2 detect pantograph, outer triggering signal is sent to the first industry
The industry flash of light of camera C1 and first lamp group L1-1, the second industry flash of light lamp group L1-2, work at the same time the two；When the second photoelectric transfer
When sensor P2 detects pantograph, send outer triggering signal to the second industrial camera C2 and third industry flash of light lamp group L2-1,
4th industry flash of light lamp group L2-2, works at the same time the two, and close number identification device AEI.
A kind of municipal rail train pantograph goat's horn monitoring method, comprising the following steps:
Step 1, initial data obtain: the first industrial camera C1 obtains pantograph pan leading flank image, the second industrial phase
Machine C2 obtains pantograph pan trailing flank image, and number identification device AEI obtains the vehicle number information of current detected train；
Step 2, image filtering processing: place is filtered using Gaussian filter algorithm to the pantograph image that step 1 obtains
Reason；
Step 3, goat's horn coarse positioning: the image zooming-out pantograph pan or so that step 2 is obtained according to endpoint template matching
Extreme coordinates, and coarse positioning is carried out to the image that step 2 obtains according to extreme coordinates；
Step 4, goat's horn fine positioning: the goat's horn image after the coarse positioning that goat's horn template and step 3 are obtained carries out template
Match, realizes goat's horn fine positioning；
Step 5, goat's horn basis of characterization differentiate: by the goat's horn basis of characterization based on edge direction, the essence that step 4 is obtained
Goat's horn image after positioning is further identified, the final result whether pantograph goat's horn lacks damage is obtained.
Further, goat's horn coarse positioning described in step 3, specific as follows:
Step 3.1, according to the feature of pantograph pan or so endpoint, pantograph pan is obtained by endpoint template matching
Left and right extreme coordinates；
Step 3.2, according to the resulting left and right extreme coordinates of step 3.1, in conjunction with the tilt angle of pantograph at runtime, if
Determine goat's horn shared area size in the picture, and interception obtains coarse positioning goat's horn image from the image that step 2 obtains.
Further, goat's horn fine positioning described in step 4, specific as follows:
Step 4.1, from the goat's horn image after the coarse positioning that step 3 obtains, artificial selection is several to complete goat's horn image,
It is fabricated to goat's horn template；
Step 4.2, by the left goat's horn image of coarse positioning and left goat's horn template image binaryzation, and by left goat's horn template two
Translation obtains the receptance function value of corresponding position, receptance function in the left goat's horn image of value are as follows:
In formula: RL(il, jl) is left goat's horn image current location receptance function value；Nl, ml are left goat's horn template image
Line number and columns；Il≤Nl-nl, jl≤Ml-ml, Nl, Ml are the line number and columns of left goat's horn image；Eml(iml, jml) is a left side
The value that iml row jml is arranged in goat's horn template binary image；El(il+iml-1, jl+jml-1) is left goat's horn image binaryzation figure
The value of il+iml-1 row jml-1 column as in；
By taking maximum row and column in receptance function value, it is set as RL(ilmax,jlmax), the horizontal seat of left goat's horn can be obtained
Mark translation variable tlxFor jlmax- 1, ordinate translates variable tlyFor ilmax-1；Left goat's horn template is obtained through translation transformation:
In formula, iml1、jml1Line position after respectively left goat's horn template translation transformation is set and column position；
Step 4.3 obtains the left goat's horn template progress rotation transformation after translation:
In formula: iml2、jml2Line position after respectively left goat's horn template rotation transformation is set and column position, and α is rotation angle；
Left goat's horn template after conversion is put into image identical with goat's horn figure ranks number, and keeps corresponding position not
Become, obtains postrotational receptance function RL(α) are as follows:
By the way that the value of α is varied multiple times, the corresponding angle [alpha] of peak response functional value is obtainedmax, and by αmaxFinally it is set to goat's horn
Rotation angle needed for fine positioning finally determines left goat's horn fine positioning template are as follows:
The determination method of right goat's horn fine positioning template is same as above.
Further, goat's horn basis of characterization described in step 5 differentiates, specific as follows:
One step 5.1, setting receptance function value Rmin, and by itself and peak response functional value ratio required by step 4.3
Compared with if the peak response functional value R detectedLLess than Rmin, then this left goat's horn missing is determined, without carrying out subsequent identification；If prison
The peak response functional value R measuredLGreater than Rmin, then enter and identify in next step；
Step 5.2, the respectively template image to left goat's horn and image to be monitored carry out edge detection；
The direction vector of each point is normalized step 5.3, is then used as similarity measure using formula (6):
In formula: (xmk,ymk) be template edge point direction vector；(xk,yk) be image to be detected marginal point direction to
Amount；N is the number of template edge point；M is the number of image to be detected marginal point；(imk,jmk) be template edge point coordinate；
(ik,jk) be image to be detected marginal point coordinate；
The integrated degree of left goat's horn to be identified can be determined by similarity measure S: when similarity measure S is less than the phase of setting
Likelihood metric threshold value SminWhen, then determine left goat's horn missing, and alarm；When similarity measure S is greater than similarity measure threshold value SminWhen,
Then determine left goat's horn can normal use, and estimate the service life of left goat's horn；
The determination method of right goat's horn missing is same as above.
Compared with prior art, the present invention its remarkable advantage are as follows: (1) can be realized the contactless monitoring to goat's horn, have
There is monitoring efficiency height, have the advantages of safety guarantee to staff；(2) coarse positioning and fine positioning are carried out to pantograph goat's horn, mentioned
The high positioning accuracy of pantograph goat's horn；(3) using the similarity measure based on edge direction as goat's horn basis of characterization, and
The direction vector of marginal point is normalized when calculating similarity measure, effectively avoids illumination variation to monitoring result
Influence.
Present invention is further described in detail with reference to the accompanying drawing.
Detailed description of the invention
Fig. 1 is the flow chart of municipal rail train pantograph goat's horn monitoring method of the present invention.
Fig. 2 is that municipal rail train pantograph goat's horn monitors system data acquisition schematic diagram in the present invention, wherein (a) is to face
Figure (b) is top view.
Fig. 3 is pantograph original graph in the embodiment of the present invention.
Fig. 4 is pantograph or so goat's horn Prototype drawing in the embodiment of the present invention, wherein (a) is left goat's horn Prototype drawing, it is (b) right side
Goat's horn Prototype drawing.
Fig. 5 is pantograph goat's horn coarse positioning result figure in the embodiment of the present invention.
Fig. 6 is the left goat's horn fine positioning result figure of pantograph in the embodiment of the present invention, wherein (a) is left goat's horn fine positioning result
Figure (b) is right goat's horn fine positioning result figure.
Specific embodiment
Municipal rail train pantograph goat's horn monitoring device of the present invention, including it is field control unit, field data acquisition unit, remote
Journey transmission unit, data analysis unit, data storage and release unit, client access unit, in which: field control unit,
It detects the arrival of train and obtains train vehicle number information；Field data acquisition unit, the acquisition for pantograph original image；Far
Journey transmission unit, on site with the data in equipment room transmit；Data analysis unit handles the image received；
Data storage and release unit, storage and publication for data；Client access unit, for remotely accessing.
In conjunction with Fig. 2 (a)~(b), the field control unit in municipal rail train pantograph goat's horn monitoring device of the present invention includes
PLC, wheel axial sensor D, number identification device AEI, in which: the wheel axial sensor D is mounted on track forefront
Inside, for detecting the arrival of municipal rail train and generating corresponding signal；PLC is used to receive the letter of wheel axial sensor D transmission
Number to control number identification device AEI；Number identification device AEI obtains the license number of municipal rail train by Train number recognition antenna.
Field data acquisition unit includes the first industry flash of light lamp group L1-1, the second industry flash of light lamp group L1-2, third work
Industry is glistened lamp group L2-1, the 4th industry flash of light lamp group L2-2, the first photoelectric sensor P1-1, the second photoelectric sensor P1-2, the
Three photoelectric sensor P2-1, the 4th photoelectric sensor P2-2, the first industrial camera C1, the second industrial camera C2；First light
Electric transducer P1-1 and the second photoelectric sensor P1-2 is symmetrically arranged on the top of two siding tracks, third photoelectric sensor P2-1 with
4th photoelectric sensor P2-2 is symmetrically arranged on the top of two siding tracks, and when each photoelectric sensor height and pantograph rising bow
Height is consistent where slide plate, for the industry flash of light of external trigger the first~bis- industrial camera C1, C2 and the first~tetra- lamp group L1-1,
L1-2,L2-1,L2-2；The first industry industry flash of light of flash of light lamp group L1-1 and second lamp group L1-2 is respectively arranged in two siding tracks
Top, the third industry industry flash of light of flash of light lamp group L2-1 and the 4th lamp group L2-2 are respectively arranged in the top of two siding tracks, industry
Lamp group of glistening provides light filling for the shooting of the first industrial camera C1, the second industrial camera C2；First industrial camera C1, the second industry
Camera C2 is center line camera, is installed on right above track centerline, for acquiring image.
The collecting flowchart of the field control unit and field data acquisition unit are as follows: when wheel axial sensor D is examined
When measuring train arrival, number identification device AEI is opened；When train logo passes through number identification device AEI, Train number recognition dress
The vehicle number information that AEI reads the train is set, and data analysis unit is transmitted to by remote transmission unit；When the first photoelectric sensing
When device P1-1 and the second photoelectric sensor P1-2 detect pantograph, outer triggering signal is sent to the first industrial camera C1 and the
One industry flash of light lamp group L1-1, the second industry flash of light lamp group L1-2, work at the same time the two；When the second photoelectric sensor P2 is detected
When to pantograph, sends outer triggering signal and dodged to the second industrial camera C2 and third industry flash of light lamp group L2-1, the 4th industry
Light lamp group L2-2 works at the same time the two, and closes number identification device AEI.
A kind of municipal rail train pantograph goat's horn monitoring method, comprising the following steps:
Step 1, initial data obtain: the first industrial camera C1 obtains pantograph pan leading flank image, the second industrial phase
Machine C2 obtains pantograph pan trailing flank image, and number identification device AEI obtains the vehicle number information of current detected train；
Step 2, image filtering processing: place is filtered using Gaussian filter algorithm to the pantograph image that step 1 obtains
Reason；
Step 3, goat's horn coarse positioning: the image zooming-out pantograph pan or so that step 2 is obtained according to endpoint template matching
Extreme coordinates, and coarse positioning is carried out to the image that step 2 obtains according to extreme coordinates；
Step 4, goat's horn fine positioning: the goat's horn image after the coarse positioning that goat's horn template and step 3 are obtained carries out template
Match, realizes goat's horn fine positioning；
Step 5, goat's horn basis of characterization differentiate: by the goat's horn basis of characterization based on edge direction, the essence that step 4 is obtained
Goat's horn image after positioning is further identified, the final result whether pantograph goat's horn lacks damage is obtained.
Further, goat's horn coarse positioning described in step 3, the image obtained to step 2 is according to slide plate or so endpoint goat's horn
Actual size carry out goat's horn interception, obtain coarse positioning goat's horn image, it is specific as follows:
Step 3.1, according to the feature of pantograph pan or so endpoint, pantograph pan is obtained by endpoint template matching
Left and right extreme coordinates；
Step 3.2, according to the resulting left and right extreme coordinates of step 3.1, in conjunction with the tilt angle of pantograph at runtime, if
Determine goat's horn shared area size in the picture, and interception obtains coarse positioning goat's horn image from the image that step 2 obtains.
Further, goat's horn fine positioning described in step 4 obtains more accurate goat's horn position from the goat's horn image of coarse positioning
It sets, is illustrated by taking the fine positioning of left goat's horn as an example, specific as follows:
Step 4.1, from the goat's horn image after the coarse positioning that step 3 obtains, artificial selection is several to complete goat's horn image,
It is fabricated to goat's horn template；
Step 4.2, by the left goat's horn image of coarse positioning and left goat's horn template image binaryzation, and by left goat's horn template two
Translation obtains the receptance function value of corresponding position, receptance function in the left goat's horn image of value are as follows:
In formula: RL(il, jl) is left goat's horn image current location receptance function value；Nl, ml are left goat's horn template image
Line number and columns；Il≤Nl-nl, jl≤Ml-ml, Nl, Ml are the line number and columns of left goat's horn image；Eml(iml, jml) is a left side
The value that iml row jml is arranged in goat's horn template binary image；El(il+iml-1, jl+jml-1) is left goat's horn image binaryzation figure
The value of il+iml-1 row jml-1 column as in；
By taking maximum row and column in receptance function value, it is set as RL(ilmax,jlmax), the horizontal seat of left goat's horn can be obtained
Mark translation variable tlxFor jlmax- 1, ordinate translates variable tlyFor ilmax-1；Left goat's horn template is obtained through translation transformation:
In formula, iml1、jml1Line position after respectively left goat's horn template translation transformation is set and column position；
Step 4.3 obtains the left goat's horn template progress rotation transformation after translation:
In formula: iml2、jml2Line position after respectively left goat's horn template rotation transformation is set and column position, and α is rotation angle；
Left goat's horn template after conversion is put into image identical with goat's horn figure ranks number, and keeps corresponding position not
Become, obtains postrotational receptance function RL(α) are as follows:
By the way that the value of α is varied multiple times, the corresponding angle [alpha] of peak response functional value is obtainedmax, and by αmaxFinally it is set to goat's horn
Rotation angle needed for fine positioning finally determines left goat's horn fine positioning template are as follows:
The determination method of right goat's horn fine positioning template is same as above.
Further, goat's horn basis of characterization described in step 5 differentiate, be it is a kind of based on edge direction similarity measure according to
According to, this method is free from the influence of the external environment, it is illustrated by taking left goat's horn as an example, the specific method is as follows:
One step 5.1, setting receptance function value Rmin, and by itself and peak response functional value ratio required by step 4.3
Compared with if the peak response functional value R detectedLLess than Rmin, then this left goat's horn centainly lacks, without carrying out subsequent identification；If prison
The peak response functional value R measuredLGreater than Rmin, then enter and identify in next step；
Step 5.2, the respectively template image to left goat's horn and image to be monitored carry out edge detection.
The direction vector of each point is normalized step 5.3, is then used as similarity measure using formula (6).
In formula: (xmk,ymk) be template edge point direction vector；(xk,yk) be image to be detected marginal point direction to
Amount；N is the number of template edge point；M is the number of image to be detected marginal point；(imk,jmk) be template edge point coordinate；
(ik,jk) be image to be detected marginal point coordinate.
The integrated degree of left goat's horn to be identified can be determined by similarity measure S.It is manually set when similarity measure S is less than
Similarity measure threshold value SminWhen, then determine left goat's horn missing, and alarm；When similarity measure S is greater than similarity measure threshold value
SminWhen, then determine left goat's horn can normal use, and the service life of left goat's horn can be estimated；
The determination method of right goat's horn missing is same as above.
Embodiment 1
Municipal rail train pantograph goat's horn monitoring device in the present embodiment, including field control unit, on-site data gathering list
Member, remote transmission unit, data analysis unit, data storage and release unit, client access unit.Wherein, field control
The arrival of unit detection train simultaneously obtains train vehicle number information；Field data acquisition unit is adopted for pantograph original image
Collection；Remote transmission unit is used to be transmitted with the data in equipment room on site；Data analysis unit to the image received at
Reason；Data storage is used for the storage and publication of data with release unit；Client access unit is for remotely accessing.
In conjunction with Fig. 2 (a)~(b), field control unit includes PLC, wheel axial sensor D, number identification device AEI,
Wherein:
The wheel axial sensor D is mounted on the inside of track forefront, for detecting arrival and the generation of municipal rail train
Corresponding signal；PLC is used to receive the signal of wheel axial sensor D transmission to control number identification device AEI；Train number recognition dress
Set the license number that AEI obtains municipal rail train by Train number recognition antenna.
Field data acquisition unit includes the first industry flash of light lamp group L1-1, the second industry flash of light lamp group L1-2, third work
Industry is glistened lamp group L2-1, the 4th industry flash of light lamp group L2-2, the first photoelectric sensor P1-1, the second photoelectric sensor P1-2, the
Three photoelectric sensor P2-1, the 4th photoelectric sensor P2-2, the first industrial camera C1, the second industrial camera C2；First light
Electric transducer P1-1 and the second photoelectric sensor P1-2 is symmetrically arranged on the top of two siding tracks, third photoelectric sensor P2-1 with
4th photoelectric sensor P2-2 is symmetrically arranged on the top of two siding tracks, and when each photoelectric sensor height and pantograph rising bow
Height is consistent where slide plate, for the industry flash of light of external trigger the first~bis- industrial camera C1, C2 and the first~tetra- lamp group L1-1,
L1-2,L2-1,L2-2；The first industry industry flash of light of flash of light lamp group L1-1 and second lamp group L1-2 is respectively arranged in two siding tracks
Top, the third industry industry flash of light of flash of light lamp group L2-1 and the 4th lamp group L2-2 are respectively arranged in the top of two siding tracks, industry
Lamp group of glistening provides light filling for the shooting of the first industrial camera C1, the second industrial camera C2；First industrial camera C1, the second industry
Camera C2 is center line camera, is installed on right above track centerline, for acquiring image.
The collecting flowchart of the acquisition system are as follows: when wheel axial sensor D detects that train arrives, open license number
Identification device AEI；When train logo passes through number identification device AEI, number identification device AEI reads the license number letter of the train
Breath, and data analysis unit is transmitted to by remote transmission unit；When the first photoelectric sensor P1-1 and the second photoelectric sensor
When P1-2 detects pantograph, outer triggering signal is sent to the industry flash of light of the first industrial camera C1 and first lamp group L1-1, the
Two industry flash of light lamp group L1-2, work at the same time the two；When the second photoelectric sensor P2 detects pantograph, external touching is sent
The second industrial camera C2 and third industry flash of light lamp group L2-1, the 4th industry flash of light lamp group L2-2 are signaled to, makes the two simultaneously
Work, and close number identification device AEI.
Municipal rail train pantograph goat's horn monitoring method in the present embodiment, comprising the following steps:
Step 1, initial data obtain: initial data is obtained by positive and negative industrial camera and number identification device acquisition.First
Industrial camera C1 obtains pantograph pan leading flank image, and the second industrial camera C2 obtains pantograph pan trailing flank image, vehicle
Number identification device AEI obtains the vehicle number information of current detected train.
Step 2, image filtering processing: the pantograph image of positive and negative camera acquisition is filtered using Gaussian filter algorithm
Processing；
Step 3, goat's horn coarse positioning: the image zooming-out pantograph pan or so that step 2 is obtained according to endpoint template matching
Extreme coordinates, and coarse positioning is carried out to the image that step 2 obtains according to extreme coordinates, the specific steps are as follows:
(3.1) according to the feature of pantograph pan or so endpoint, a left side for pantograph pan is obtained by endpoint template matching
Right endpoint coordinate.
(3.2) according to the resulting left and right extreme coordinates of step (3.1), in conjunction with the tilt angle of pantograph at runtime, if
Determine goat's horn shared area size in the picture, and interception obtains coarse positioning goat's horn image from the image that step 2 obtains.
Step 4, goat's horn fine positioning: template is carried out by the goat's horn image that the goat's horn template and step 3 of engineer obtain
Goat's horn fine positioning is realized in matching.Specific step is as follows:
(4.1) from the goat's horn image after the coarse positioning that step 3 obtains, artificial selection is several to complete goat's horn image, system
It is made goat's horn template；
(4.2) by the left goat's horn image of coarse positioning and left goat's horn template image binaryzation, and by left goat's horn template in two-value
Translation obtains the receptance function value of corresponding position, receptance function in the left goat's horn image changed are as follows:
In formula: RL(il, jl) is left goat's horn image current location receptance function value；Nl, ml are left goat's horn template image
Line number and columns；Il≤Nl-nl, jl≤Ml-ml, Nl, Ml are the line number and columns of left goat's horn image；Eml(iml, jml) is a left side
The value that iml row jml is arranged in goat's horn template binary image；El(il+iml-1, jl+jml-1) is left goat's horn image binaryzation figure
The value of il+iml-1 row jml-1 column as in；
By taking maximum row and column in receptance function value, it is set as RL(ilmax,jlmax), the horizontal seat of left goat's horn can be obtained
Mark translation variable tlxFor jlmax- 1, ordinate translates variable tlyFor ilmax-1；Left goat's horn template is obtained through translation transformation:
In formula, iml1、jml1Line position after respectively left goat's horn template translation transformation is set and column position；
(4.3) rotation transformation is carried out to the left goat's horn template after translation to obtain:
In formula: iml2、jml2Line position after respectively left goat's horn template rotation transformation is set and column position, and α is rotation angle；
Left goat's horn template after conversion is put into image identical with goat's horn figure ranks number, and keeps corresponding position not
Become, obtains postrotational receptance function RL(α) are as follows:
By the way that the value of α is varied multiple times, the corresponding angle [alpha] of peak response functional value is obtainedmax, and by αmaxFinally it is set to goat's horn
Rotation angle needed for fine positioning finally determines left goat's horn fine positioning template are as follows:
The determination method of right goat's horn fine positioning template is same as above.
Step 5, goat's horn basis of characterization differentiate: the sheep obtained by the goat's horn basis of characterization based on edge direction to step 4
Angle image is further identified, the final result whether pantograph goat's horn lacks damage is obtained.It is said by taking left goat's horn as an example
Bright, the specific method is as follows:
(5.1) a receptance function value R is setmin, and by its with step 4.3 required by compared with peak response functional value, if
The peak response functional value R detectedLLess than Rmin, then this left goat's horn missing is determined, without carrying out subsequent identification；If detecting
Peak response functional value RLGreater than Rmin, then enter and identify in next step；
(5.2) template image to left goat's horn and image to be monitored carry out edge detection respectively.
(5.3) direction vector of each point is normalized, is then used as similarity measure using formula (6):
In formula: (xmk,ymk) be template edge point direction vector；(xk,yk) be image to be detected marginal point direction to
Amount；N is the number of template edge point；M is the number of image to be detected marginal point；(imk,jmk) be template edge point coordinate；
(ik,jk) be image to be detected marginal point coordinate.
The integrated degree of left goat's horn to be identified can be determined by similarity measure S: when similarity measure S is less than the phase of setting
Likelihood metric threshold value SminWhen, then determine left goat's horn missing, and alarm；When similarity measure S is greater than similarity measure threshold value SminWhen,
Then determine left goat's horn can normal use, and estimate the service life of left goat's horn.
The determination method of right goat's horn missing is same as above.
According to the data capture method of step 1, pantograph original image as shown in Figure 3 is obtained.By step 2 to image
Gaussian filtering pretreatment is carried out, the image for filtering out partial noise is obtained.
It is obtained using such as Fig. 4 (a)~(b) goat's horn template image according to endpoint template matching according to the method in step 3
Pantograph pan or so extreme coordinates, and coarse positioning is carried out to the image that pretreatment obtains according to extreme coordinates and is obtained such as Fig. 5 institute
Show.
According to the method in step 4, left goat's horn image and left goat's horn template image binary value E are obtainedl(il+iml-1,
jl+jml-1)、Eml(iml, jml), image response functional value RL(il, jl) and by translating postrotational receptance function RL(α)。
The corresponding angle [alpha] of peak response functional value can be obtained in value by the way that α is varied multiple timesmax, finally determine left goat's horn fine positioning image such as
Shown in Fig. 6 (a)~(b).
According to the method in step 5, after mass data is tested, the receptance function value threshold value R of artificial settingsminFor
7000, the similarity measure threshold value S of artificial settingsminIt is 0.6.Step 4 is obtained by the goat's horn basis of characterization based on edge direction
Goat's horn image carry out further identification and obtain similarity measure S, and differentiated to obtain to whether pantograph goat's horn lacks according to S
Final result.
In summary, the present invention can be realized the contactless monitoring to goat's horn, have monitoring efficiency height, precision high, right
Staff has the advantages of safety guarantee.Meanwhile effectively avoiding influence of the illumination variation to monitoring result.
Claims (8)
1. a kind of municipal rail train pantograph goat's horn monitoring device, which is characterized in that including field control unit, on-site data gathering
Unit, remote transmission unit, data analysis unit, data storage and release unit, client access unit, in which:
Field control unit detects the arrival of train and obtains train vehicle number information；
Field data acquisition unit, the acquisition for pantograph original image；
Remote transmission unit, on site with the data in equipment room transmit；
Data analysis unit handles the image received；
Data storage and release unit, storage and publication for data；
Client access unit, for remotely accessing.
2. municipal rail train pantograph goat's horn monitoring device according to claim 1, which is characterized in that the field control
Unit includes PLC, wheel axial sensor (D), number identification device (AEI), in which:
The wheel axial sensor (D) is mounted on the inside of track forefront, for detecting the arrival of municipal rail train and generating phase
Induction signal；
PLC is used to receive the signal of wheel axial sensor (D) transmission to control number identification device (AEI)；
Number identification device (AEI) obtains the license number of municipal rail train by Train number recognition antenna.
3. municipal rail train pantograph goat's horn monitoring device according to claim 1, which is characterized in that the field data
Acquisition unit includes the first industry flash of light lamp group (L1-1), the second industry flash of light lamp group (L1-2), third industry flash of light lamp group
(L2-1), the 4th industry flash of light lamp group (L2-2), the first photoelectric sensor (P1-1), the second photoelectric sensor (P1-2), third
Photoelectric sensor (P2-1), the 4th photoelectric sensor (P2-2), the first industrial camera (C1), the second industrial camera (C2)；
First photoelectric sensor (P1-1) and the second photoelectric sensor (P1-2) are symmetrically arranged on the top of two siding tracks, the
Three photoelectric sensors (P2-1) and the 4th photoelectric sensor (P2-2) are symmetrically arranged on the top of two siding tracks, and each photoelectric sensing
Height is consistent where device height and slide plate when pantograph rising bow, for the first~bis- industrial camera of external trigger (C1, C2) and
The first~tetra- industry flash of light lamp group (L1-1, L1-2, L2-1, L2-2)；First industry flash of light lamp group (L1-1) is dodged with the second industry
Light lamp group (L1-2) is respectively arranged in the top of two siding tracks, lamp group that third industry is glistened lamp group (L2-1) and the 4th industry is glistened
(L2-2) it is respectively arranged in the top of two siding tracks, industry flash of light lamp group is the first industrial camera (C1), the second industrial camera
(C2) shooting provides light filling；First industrial camera (C1), the second industrial camera (C2) are center line camera, are installed in track
Right above heart line, for acquiring image.
4. municipal rail train pantograph goat's horn monitoring device according to claim 2 or 3, which is characterized in that the scene
The collecting flowchart of control unit and field data acquisition unit are as follows: when wheel axial sensor (D) detects that train arrives, open
Open number identification device (AEI)；When train logo passes through number identification device (AEI), number identification device (AEI) is read should
The vehicle number information of train, and data analysis unit is transmitted to by remote transmission unit；When the first photoelectric sensor (P1-1) with
When second photoelectric sensor (P1-2) detects pantograph, outer triggering signal is sent to the first industrial camera (C1) and the first work
Industry is glistened lamp group (L1-1), the second industry is glistened lamp group (L1-2), works at the same time the two；When the second photoelectric sensor (P2) is examined
When measuring pantograph, outer triggering signal is sent to the second industrial camera (C2) and third industry flash of light lamp group (L2-1), the 4th
Industry flash of light lamp group (L2-2) works at the same time the two, and closes number identification device (AEI).
5. a kind of municipal rail train pantograph goat's horn monitoring method, which comprises the following steps:
Step 1, initial data obtain: the first industrial camera (C1) obtains pantograph pan leading flank image, the second industrial camera
(C2) pantograph pan trailing flank image is obtained, number identification device (AEI) obtains the vehicle number information of current detected train；
Step 2, image filtering processing: the pantograph image obtained to step 1 is filtered using Gaussian filter algorithm；
Step 3, goat's horn coarse positioning: image zooming-out pantograph pan that step 2 is obtained according to endpoint template matching or so endpoint
Coordinate, and coarse positioning is carried out to the image that step 2 obtains according to extreme coordinates；
Step 4, goat's horn fine positioning: the goat's horn image after the coarse positioning that goat's horn template and step 3 are obtained carries out template matching, real
Existing goat's horn fine positioning；
Step 5, goat's horn basis of characterization differentiate: by the goat's horn basis of characterization based on edge direction, the fine positioning that step 4 is obtained
Goat's horn image afterwards is further identified, the final result whether pantograph goat's horn lacks damage is obtained.
6. municipal rail train pantograph goat's horn monitoring method according to claim 5, which is characterized in that sheep described in step 3
Angle coarse positioning, specific as follows:
Step 3.1, according to the feature of pantograph pan or so endpoint, the left and right of pantograph pan is obtained by endpoint template matching
Extreme coordinates；
Step 3.2 sets sheep in conjunction with the tilt angle of pantograph at runtime according to the resulting left and right extreme coordinates of step 3.1
Angle shared area size in the picture, and interception obtains coarse positioning goat's horn image from the image that step 2 obtains.
7. municipal rail train pantograph goat's horn monitoring method according to claim 5 or 6, which is characterized in that described in step 4
Goat's horn fine positioning, specific as follows:
Step 4.1, from the goat's horn image after the coarse positioning that step 3 obtains, artificial selection is several to complete goat's horn image, production
At goat's horn template；
Step 4.2, by the left goat's horn image of coarse positioning and left goat's horn template image binaryzation, and by left goat's horn template in binaryzation
Left goat's horn image in translation obtain the receptance function value of corresponding position, receptance function are as follows:
In formula: RL(il, jl) is left goat's horn image current location receptance function value；Nl, ml be left goat's horn template image line number and
Columns；Il≤Nl-nl, jl≤Ml-ml, Nl, Ml are the line number and columns of left goat's horn image；Eml(iml, jml) is left goat's horn mould
The value that iml row jml is arranged in plate binary image；El(il+iml-1, jl+jml-1) is il+ in left goat's horn image binaryzation image
The value of iml-1 row jml-1 column；
By taking maximum row and column in receptance function value, it is set as RL(ilmax,jlmax), it is flat that left goat's horn abscissa can be obtained
Move variable tlxFor jlmax- 1, ordinate translates variable tlyFor ilmax-1；Left goat's horn template is obtained through translation transformation:
In formula, iml1、jml1Line position after respectively left goat's horn template translation transformation is set and column position；
Step 4.3 obtains the left goat's horn template progress rotation transformation after translation:
In formula: iml2、jml2Line position after respectively left goat's horn template rotation transformation is set and column position, and α is rotation angle；
Left goat's horn template after conversion is put into image identical with goat's horn figure ranks number, and keeps corresponding position constant, is obtained
To postrotational receptance function RL(α) are as follows:
By the way that the value of α is varied multiple times, the corresponding angle [alpha] of peak response functional value is obtainedmax, and by αmaxIt is fixed to be finally set to goat's horn essence
Rotation angle needed for position, finally determines left goat's horn fine positioning template are as follows:
The determination method of right goat's horn fine positioning template is same as above.
8. municipal rail train pantograph goat's horn on-line monitoring method according to claim 7, which is characterized in that described in step 5
Goat's horn basis of characterization differentiate, it is specific as follows:
One step 5.1, setting receptance function value Rmin, and by its with step 4.3 required by compared with peak response functional value, if
The peak response functional value R detectedLLess than Rmin, then this left goat's horn missing is determined, without carrying out subsequent identification；If detecting
Peak response functional value RLGreater than Rmin, then enter and identify in next step；
Step 5.2, the respectively template image to left goat's horn and image to be monitored carry out edge detection；
The direction vector of each point is normalized step 5.3, is then used as similarity measure using formula (6):
In formula: (xmk,ymk) be template edge point direction vector；(xk,yk) be image to be detected marginal point direction vector；n
For the number of template edge point；M is the number of image to be detected marginal point；(imk,jmk) be template edge point coordinate；(ik,
jk) be image to be detected marginal point coordinate；
The integrated degree of left goat's horn to be identified can be determined by similarity measure S: when similarity measure S is less than the similarity of setting
Measure threshold value SminWhen, then determine left goat's horn missing, and alarm；When similarity measure S is greater than similarity measure threshold value SminWhen, then sentence
Fixed left goat's horn can normal use, and estimate the service life of left goat's horn；
The determination method of right goat's horn missing is same as above.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810424208.4A CN109029539A (en) | 2018-05-07 | 2018-05-07 | A kind of municipal rail train pantograph goat's horn monitoring device and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810424208.4A CN109029539A (en) | 2018-05-07 | 2018-05-07 | A kind of municipal rail train pantograph goat's horn monitoring device and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109029539A true CN109029539A (en) | 2018-12-18 |
Family
ID=64611564
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810424208.4A Withdrawn CN109029539A (en) | 2018-05-07 | 2018-05-07 | A kind of municipal rail train pantograph goat's horn monitoring device and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109029539A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110455214A (en) * | 2019-07-30 | 2019-11-15 | 中车青岛四方机车车辆股份有限公司 | A kind of abrasion of pantograph pan condition monitoring system and method |
| CN111260629A (en) * | 2020-01-16 | 2020-06-09 | 成都地铁运营有限公司 | Pantograph structure abnormity detection algorithm based on image processing |
-
2018
- 2018-05-07 CN CN201810424208.4A patent/CN109029539A/en not_active Withdrawn
Non-Patent Citations (1)
| Title |
|---|
| 陈双: "基于图像处理的受电弓故障检测算法研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110455214A (en) * | 2019-07-30 | 2019-11-15 | 中车青岛四方机车车辆股份有限公司 | A kind of abrasion of pantograph pan condition monitoring system and method |
| CN111260629A (en) * | 2020-01-16 | 2020-06-09 | 成都地铁运营有限公司 | Pantograph structure abnormity detection algorithm based on image processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109658397A (en) | A kind of rail polling method and system | |
| CN202793389U (en) | On-line routing inspection device for high-speed rail overhead line system | |
| CN103837087B (en) | Pantograph automatic testing method based on active shape model | |
| CN111524365B (en) | Method for classifying vehicle types by using multiple geomagnetic sensors | |
| CN101833099B (en) | Locked tracking method of helicopter in the inspection process of electric power circuits | |
| CN107393270A (en) | A kind of portable vision inspection device and method for electric meter detection | |
| WO2018122588A1 (en) | Method for detecting pedestrian traffic by using wi-fi probe | |
| CN109029539A (en) | A kind of municipal rail train pantograph goat's horn monitoring device and method | |
| CN105975979A (en) | Instrument detection method based on machine vision | |
| CN207850304U (en) | A kind of electrification railway contact net detecting system | |
| CN101281143A (en) | System and method for safe detecting chassis of automobile | |
| CN108981565A (en) | A kind of municipal rail train pantograph disalignment monitoring device and method | |
| CN108680104A (en) | A kind of municipal rail train bow failure monitoring device and method | |
| CN205157546U (en) | Air quality test equipment with quality class appraises function | |
| CN111565024A (en) | Solar cell panel fault detection and operation efficiency prediction device and method | |
| CN110427943A (en) | A kind of intelligent electric meter technique for partitioning based on R-CNN | |
| CN110435689B (en) | Detection method of intelligent obstacle detection vehicle based on modular control | |
| CN1908987A (en) | Method for diagnosing accidents of waveshape recognition based substation locomotive traction circuit | |
| CN102592473A (en) | Vehicle recognition method and vehicle recognition system | |
| CN201203581Y (en) | Safety detection system for chassis of automobile | |
| CN211869371U (en) | Intelligent obstacle detection vehicle based on modular control | |
| CN1363728A (en) | Method and device for automatically measuring distribution of anode current in Al electrolyzer | |
| CN201717992U (en) | Video passenger detector | |
| CN208335321U (en) | A kind of gate inhibition's recognition of face all-in-one machine | |
| CN108839676B (en) | Online dynamic measurement device and measurement method for geometric parameters of train wheels |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20181218 |