CN110044645A - Protective cover board for contact rails condition detecting system and method - Google Patents
Protective cover board for contact rails condition detecting system and method Download PDFInfo
- Publication number
- CN110044645A CN110044645A CN201910375159.4A CN201910375159A CN110044645A CN 110044645 A CN110044645 A CN 110044645A CN 201910375159 A CN201910375159 A CN 201910375159A CN 110044645 A CN110044645 A CN 110044645A
- Authority
- CN
- China
- Prior art keywords
- protective cover
- cover board
- contact rails
- data
- image data
- 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
- 230000001681 protective effect Effects 0.000 title claims abstract description 301
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000001514 detection method Methods 0.000 claims abstract description 65
- 230000007547 defect Effects 0.000 claims abstract description 41
- 239000004020 conductor Substances 0.000 claims description 40
- 238000003062 neural network model Methods 0.000 claims description 31
- 230000008569 process Effects 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 11
- 238000007689 inspection Methods 0.000 claims description 8
- 208000037656 Respiratory Sounds Diseases 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 210000005036 nerve Anatomy 0.000 claims description 2
- 238000005259 measurement Methods 0.000 abstract description 7
- 238000010586 diagram Methods 0.000 description 13
- 230000006870 function Effects 0.000 description 8
- 238000009434 installation Methods 0.000 description 8
- 230000008859 change Effects 0.000 description 6
- 230000033001 locomotion Effects 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000006073 displacement reaction Methods 0.000 description 4
- 238000003384 imaging method Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 201000010099 disease Diseases 0.000 description 3
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000002372 labelling Methods 0.000 description 3
- 206010014357 Electric shock Diseases 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000009413 insulation Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 210000001367 artery Anatomy 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 230000034994 death Effects 0.000 description 1
- 231100000517 death Toxicity 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- JEIPFZHSYJVQDO-UHFFFAOYSA-N ferric oxide Chemical compound O=[Fe]O[Fe]=O JEIPFZHSYJVQDO-UHFFFAOYSA-N 0.000 description 1
- 238000013100 final test Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000000178 monomer Substances 0.000 description 1
- 210000004218 nerve net Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012634 optical imaging Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 210000003462 vein Anatomy 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B61—RAILWAYS
- B61K—AUXILIARY EQUIPMENT SPECIALLY ADAPTED FOR RAILWAYS, NOT OTHERWISE PROVIDED FOR
- B61K9/00—Railway vehicle profile gauges; Detecting or indicating overheating of components; Apparatus on locomotives or cars to indicate bad track sections; General design of track recording vehicles
- B61K9/08—Measuring installations for surveying permanent way
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M17/00—Testing of vehicles
- G01M17/08—Railway vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The present invention provides a kind of protective cover board for contact rails condition detecting system and methods, this method comprises: data acquisition unit, positioning unit and detection unit, wherein, data acquisition unit is installed on operation vehicle car body, it is connect with detection unit, for acquiring the status data of protective cover board for contact rails, and the status data of protective cover board for contact rails is sent to detection unit when runing vehicle operation;Positioning unit is installed on operation vehicle vehicle bottom, connect with detection unit, and for runing the location information for obtaining protective cover board for contact rails when vehicle operation, the location information of protective cover board for contact rails is sent to detection unit;Detection unit obtains the defect state information of protective cover board for contact rails for the status data of protective cover board for contact rails based on the received and the location information of protective cover board for contact rails.The present invention can detect the state of protective cover board for contact rails under normal driving condition, and detection speed is fast, and measurement accuracy is high.
Description
Technical field
The present invention relates to power supply of urban orbit traffic conductor rail detection technique field more particularly to a kind of protective cover board for contact rails
Condition detecting system and method.
Background technique
Conductor rail is the spy for transferring electrical energy into subway and City Rail Transit System electric traction vehicle (electric EMU)
Different power supply unit.Conductor rail arranges that electric EMU is contacted by the collector shoe stretched out and receives electric energy along rail.
Conductor rail power supply mode be divided into again lower part award streaming conductor rail power supply, top award streaming conductor rail power supply and side award
The power supply of streaming conductor rail.It is DC1500V that the general voltage of streaming conductor rail is awarded in lower part, is mounted with insulation protective hood, energy at an upper portion thereof
It enough avoids the risk of electric shock and such as rains with bad weather, dust, snows and award the influence of stream to conductor rail power supply, therefore many cities
The power supply of streaming conductor rail is awarded using train lower part, as depicted in figs. 1 and 2, gives shield 5, conductor rail 6 and walking rail 7
Positional relationship.
As depicted in figs. 1 and 2, the perforation of shield 5 is overlapped in 6 top of conductor rail, generally lacks following formula design using rectangular,
Exposed conductor rail ontology is protected from horizontal two sides and vertical upper surface.Protective cover board for contact rails 5 is prevented by shield support card, commonly
Shield 8, bracket shield 9, electrical connection shield (use at online cable), special middle anchor shield are (at special central anchor section
Using) etc. composition.Common shield is spaced the shield support card erection installation of 500mm arrangement by being placed on conductor rail,
General installation length 4200mm.Bracket shield is mounted at anchor point insulating support, is buckled in the common of the anchor point two sides
On shield (snapping is overlapped length 200mm).Such just formation is open to traffic and completely obscured insulation protection, Fig. 3 and Fig. 4
The schematic diagram of streaming protective cover board for contact rails installation is awarded for lower part.In railway roadbed installation foundation, the relative position of insulating support 10, contact
Under the influence of many dynamic factors such as the real-time dynamic of 6 ontology of rail and the disordered motion of shield support card, shield 5 not can guarantee
Operating position is opposing stationary, should mutually overlapped shield because conductor rail is exposed caused by displacement, shield sink, even partially
, more very there is hurt crackle, there is very big security risk in the case where moving intrusion train limit.Therefore, this phenomenon is stranded always
The unit of operation for disturbing conductor rail equipment disturbs normal production maintenance order.
It is DC1500V that the general voltage of streaming conductor rail is awarded in lower part, by the way of manually inspection protective cover board for contact rails state,
Low efficiency, heavy workload are likely to result in high-voltage end ground connection and cause power failure, the even tragedy of personnel's electric shock injures and deaths.Cause
This can only patrol in limited power-off time, and inspection point is few, and density is low, and is detached from dynamic situation when train operation,
These are all unfavorable for the scientific evaluation management and planned maintenance of conductor rail equipment.Therefore, how to be detected under normal driving condition
The state of protective cover board for contact rails becomes a urgent problem needed to be solved.
Summary of the invention
The embodiment of the present invention proposes a kind of protective cover board for contact rails condition detecting system, to detect under normal driving condition
The state of protective cover board for contact rails, detection speed is fast, and measurement accuracy is high, which includes:
Data acquisition unit, positioning unit and detection unit, wherein
Data acquisition unit is installed on operation vehicle car body, is connect with detection unit, for acquiring when runing vehicle operation
The status data of protective cover board for contact rails, and the status data of protective cover board for contact rails is sent to detection unit;
Positioning unit is installed on operation vehicle vehicle bottom, connect with detection unit, for obtaining conductor rail when runing vehicle operation
The location information of protective cover board for contact rails is sent to detection unit by the location information of shield;
Detection unit, for the status data of protective cover board for contact rails based on the received and the position letter of protective cover board for contact rails
Breath, obtains the defect state information of protective cover board for contact rails.
The embodiment of the present invention proposes a kind of protective cover board for contact rails condition detection method, and this method is anti-applied to above-mentioned conductor rail
Shield condition detecting system, to detect the state of protective cover board for contact rails under normal driving condition, detection speed is fast, measurement essence
Degree is high, this method comprises:
When runing vehicle operation, the status data of protective cover board for contact rails is acquired;
Runing the location information that protective cover board for contact rails is obtained when vehicle operation;
According to the location information of the status data of protective cover board for contact rails and protective cover board for contact rails, protective cover board for contact rails is obtained
Defect state information.
The embodiment of the present invention also proposed a kind of computer equipment, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, the processor realize that above-mentioned conductor rail is anti-when executing the computer program
Shield condition detection method.
The embodiment of the present invention also proposed a kind of computer readable storage medium, the computer-readable recording medium storage
There is the computer program for executing above-mentioned protective cover board for contact rails condition detection method.
In embodiments of the present invention, data acquisition unit acquires the status number of protective cover board for contact rails when runing vehicle operation
According to the status data of entire protective cover board for contact rails can be acquired when conductor rail is to operation vehicle powered operation;It is installed on operation vehicle vehicle
The positioning unit at bottom can run the location information that protective cover board for contact rails is obtained when vehicle operation;Finally, detection unit is according to reception
Protective cover board for contact rails status data and protective cover board for contact rails location information, can directly obtain protective cover board for contact rails lack
Status information is fallen into, is not necessarily to manual inspection, therefore detection speed is fast, measurement accuracy is high.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.In the accompanying drawings:
Fig. 1 and Fig. 2 is the schematic diagram that the power supply of streaming conductor rail is awarded in train lower part;
Fig. 3 and Fig. 4 is the schematic diagram that the installation of streaming protective cover board for contact rails is awarded in lower part;
Fig. 5 is the schematic diagram of protective cover board for contact rails condition detecting system in the embodiment of the present invention;
Fig. 6 shows to be grouped as the installation site of unit in the middle part of the protective cover board for contact rails condition detecting system of the embodiment of the present invention
It is intended to;
Fig. 7 is the principle that camera model equidistantly scans protective cover board for contact rails in the embodiment of the present invention;
Fig. 8 is the specific of the defect state information that state recognition module identifies protective cover board for contact rails in the embodiment of the present invention
Process;
Fig. 9 is that the neural network model of protective cover board for contact rails in the embodiment of the present invention implements process;
Figure 10 is the form schematic diagram of the image data of protective cover board for contact rails in the embodiment of the present invention;
Figure 11 is the system annexation figure in the embodiment of the present invention when there are two data acquisition unit;
Figure 12 is the flow chart of protective cover board for contact rails condition detection method in the embodiment of the present invention;
Figure 13 is the detail flowchart for the protective cover board for contact rails condition detection method that the embodiment of the present invention proposes.
Specific embodiment
Understand in order to make the object, technical scheme and advantages of the embodiment of the invention clearer, with reference to the accompanying drawing to this hair
Bright embodiment is described in further details.Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but simultaneously
It is not as a limitation of the invention.
Operation vehicle maximum speed is up to 160km/h at present, in order to realize contact of the operation vehicle at maximum speed 160km/h
The dynamic monitoring of rail shield, the embodiment of the present invention propose a kind of protective cover board for contact rails condition detecting system.
Fig. 5 is the schematic diagram of protective cover board for contact rails condition detecting system in the embodiment of the present invention, as shown in figure 5, the system
Include:
Data acquisition unit, positioning unit and detection unit, wherein
Data acquisition unit is installed on operation vehicle car body, is connect with detection unit, for acquiring when runing vehicle operation
The status data of protective cover board for contact rails, and the status data of protective cover board for contact rails is sent to detection unit;
Positioning unit is installed on operation vehicle vehicle bottom, connect with detection unit, for obtaining conductor rail when runing vehicle operation
The location information of protective cover board for contact rails is sent to detection unit by the location information of shield;
Detection unit, for the status data of protective cover board for contact rails based on the received and the position letter of protective cover board for contact rails
Breath, obtains the defect state information of protective cover board for contact rails.
In embodiments of the present invention, data acquisition unit acquires the status number of protective cover board for contact rails when runing vehicle operation
According to the status data of entire protective cover board for contact rails can be acquired when conductor rail is to operation vehicle powered operation;It is installed on operation vehicle vehicle
The positioning unit at bottom can run the location information that protective cover board for contact rails is obtained when vehicle operation;Finally, detection unit is according to reception
Protective cover board for contact rails status data and protective cover board for contact rails location information, can directly obtain protective cover board for contact rails lack
Status information is fallen into, is not necessarily to manual inspection, therefore detection speed is fast, measurement accuracy is high.
Fig. 6 shows to be grouped as the installation site of unit in the middle part of the protective cover board for contact rails condition detecting system of the embodiment of the present invention
It is intended to, as shown in fig. 6, data acquisition unit 1 is installed on operation vehicle car body, detection unit 2 is located on operation vehicle, can also position
In in the fixed equipment of ground by wireless network transmissions data, but when detection unit 2 is located on operation vehicle, with data acquisition unit
1 communication efficiency is best;Positioning unit 4 is installed on operation vehicle vehicle bottom, obtains the location information of protective cover board for contact rails, alternatively referred to as
The mileage information of protective cover board for contact rails, positioning unit can obtain essence according to its preset electronic labeling information in internal mileage position
The location information of true protective cover board for contact rails, so that even if shield occurs mobile remain on and can collect protective cover board for contact rails
Accurate location information, the precision of the defect state information of the protective cover board for contact rails obtained from is higher.Protective cover board for contact rails
Status data and protective cover board for contact rails location information, can be sent to by cable (Cameralink cable or kilomega network)
Detection unit.
In addition, data acquisition unit Highgrade integration, miniaturization, equipment volume is small, light-weight, and detection unit can be for just
Take formula computer, the whole easy to use and flexible of rail-engaging shield condition detecting system encounters failure, direct integral replacing, i.e.,
Dress is used.
In one embodiment, the status data of protective cover board for contact rails includes in image data, pressure data and temperature data
One of which or any combination.
In the above-described embodiments, image data can be collected by optical imaging element, and pressure data can be by pressure
Sensor collects, and temperature data can be collected by temperature sensor, these may serve to analysis protective cover board for contact rails
Defect state information.Example given below is mainly by taking image data as an example.
In one embodiment, protective cover board for contact rails condition detecting system further include: it is installed on operation vehicle wheel axle box, and
The trigger unit 3 being connect with data acquisition unit, for issuing trigger signal to data acquisition unit;
Data acquisition unit is specifically used for when runing vehicle operation, based on the received trigger signal, acquisition conductor rail protection
The status data of cover, and the image data of protective cover board for contact rails is sent to detection unit.
In the above-described embodiments, trigger unit 3 is installed on operation vehicle wheel axle box, as shown in Figure 6, therefore, can be with
The operation of operation vehicle and continuously send out trigger signal, trigger signal can be issued according to certain rule, and data acquisition unit is every
It is connected to a trigger signal, carries out the state data acquisition of a protective cover board for contact rails, in the operational process of operation vehicle, installation
The status data of whole protective cover board for contact rails is just acquired in the data acquisition unit on operation vehicle car body.
When it is implemented, the structure of data acquisition unit can there are many, sufficiently to acquire the state of protective cover board for contact rails
One of embodiment is given below in data.
In one embodiment, data acquisition unit includes camera model, is used for when runing vehicle operation, according to trigger signal
Protective cover board for contact rails is scanned, the image data of protective cover board for contact rails is acquired.
In the above-described embodiments, camera can be scanned protective cover board for contact rails and take pictures, and often be connected to a trigger signal, phase
Machine carries out single pass and takes pictures, and therefore, can scan multiple images of taking pictures to whole protective cover board for contact rails, obtain through image mosaic
The image data of protective cover board for contact rails.
Camera model is using advanced high-resolution (transverse direction >=0.5mm/pix, longitudinal direction >=0.5mm/pix), low noise, small
The imaging sensor (the 4K line-scan digital camera of such as CCD, CMOS chip) of distortion, camera model can also use narrow-band-filter technology,
Sunlight interference can be effectively filtered out, picture quality is promoted.
When it is implemented, triggering is single in order to enable camera is apparent in the image of the acquisition of taking pictures to protective cover board for contact rails
First and camera can there are many fit systems, the following examples to provide one of fit system.
In one embodiment, trigger unit issues the trigger signal of fixed frequency to data acquisition unit;
Camera model is specifically used for: when runing vehicle operation, the trigger signal of fixed frequency based on the received, spatially
Equidistant scanning protective cover board for contact rails, acquires the image data of protective cover board for contact rails.
In the above-described embodiments, trigger unit issues the trigger signal of fixed frequency to data acquisition unit, so that camera
Module may be implemented equidistantly to scan protective cover board for contact rails, and Fig. 7 is that camera model equidistantly scans contact in the embodiment of the present invention
The principle of rail shield, as shown in fig. 7, firstly, the trigger signal of trigger unit sending fixed frequency, is with pulse coder
Example, pulse coder issue the pulse signal of fixed frequency, and pulse coder is a kind of rotation impulse generator, can be machine
Tool corner becomes electric pulse.Pulse coder can be with the line shaft rotation of operation vehicle, under the pulse signal of fixed frequency, arteries and veins
It rushes encoder and revolves the generated electric impulse signal sum that turns around with line shaft, i.e., all electric impulse signal sums, with fixed value K table
Show.
Using following formula, the move distance of single-revolution can be rotated in line shaft according to all electric impulse signal sums, operation vehicle,
Obtain the move distance of unit-pulse signal:
Wherein, d is the move distance of unit pulse signal;
L is to run vehicle in the move distance of line shaft rotation single-revolution
K is all electric impulse signals sum of pulse coder.
The move distance d of unit-pulse signal is also acquisition system up to minimum space resolution ratio, therefore, in operation electricity visitor
In the case that vehicle is set, when the move distance d of unit-pulse signal is bigger, i.e. pulse coder resolution ratio is higher, and image is reachable
Precision is also higher.
Using following formula, according to sample space distance, the move distance of unit-pulse signal obtains trigger pulse interval
Constant:
Wherein, m is trigger pulse interval constant;
D is required sample space distance.
It is the line scanning frequency of configurable camera according to trigger pulse interval constant, following formula can be used and obtain:
Wherein, f is the line scanning frequency of camera;
V is operating speed degree.
By formula (1)-formula (3) it is found thatIt is presented between the line scanning frequency and operating speed degree of camera linear
Transformation relation, in the case where operating speed degree is certain, the mutually confidential higher spatial resolution of acquisition needs the line of camera to scan
Frequency is bigger.When operation Chefei travels at the uniform speed, i.e. when operating speed degree V changes, need to adjust line scanning frequency f, Ji Keshi
When it is constant to show sample space distance D, and adjusting line scanning frequency f, all electric impulse signals for having no need to change pulse coder are total
Number K, it is only necessary to the width for changing unit-pulse signal, according to formula (1), the move distance d of unit-pulse signal be it is fixed,
Therefore, change the width of unit-pulse signal, i.e., adjustable line scanning frequency f, so that even if in operating speed degree V variation,
It can still keep sample space distance D constant, so that the space under the conditions of realizing non-uniform movement equidistantly samples, so that camera
The image of acquisition more evenly, i.e., it is irredundant obtain protective cover board for contact rails image data.
Other than pulse coder, trigger unit can also be using photoelectric encoder or grating encoder or radar, phase
Protection scope of the present invention should all be fallen by closing change case.
When it is implemented, data acquisition unit is to ensure that camera sufficiently acquires image, it may include light source, below for wherein
One embodiment.
In one embodiment, data acquisition unit includes light source module and Light Source Compensation module, wherein
Light source module, for illuminating protective cover board for contact rails;
Light Source Compensation module, for highlighting to the protective cover board for contact rails that light source module illuminates.
In the above-described embodiments, light source module can use LED linear light source or LVD linear light sorurce, Light Source Compensation module
Broader brightness range can be obtained, so that phase using 808nm infrared laser or 450nm blue light or 650nm feux rouges light filling
Machine has better image effect.
The needs of equidistant, high frequency time imaging samples is carried out to protective cover board for contact rails in order to meet, camera model can benefit
Be imaged with line-scan digital camera, linear array camera have the characteristics that frequency it is high, can irredundant splicing, can equidistantly be moved with short cycle
The mode of scanning is imaged.Existing operation vehicle maximum speed be 160km/h, for meet resolution ratio under the above velocity conditions≤
The sampling precision of 1mm/pixel, line-scan digital camera maximum scan working frequency need to be higher than 33.3KHz, at this time the parameter of line-scan digital camera
As shown in table 1.
The parameter of 1 line-scan digital camera of table
Camera parameter | Parameter value | Remarks |
Camera resolution | 2048pixels | |
Maximum line frequency | 80KHz | Greater than 33.3KHz |
Image output interface | GigE | GigE |
Imaging mode | The scanning of CMOS row | |
Picture dot size | 7.04um×7.04um | |
Time for exposure | 4us-30us | |
Lens interface | M42 | |
Operating voltage | 12-24V DC |
When it is implemented, the structure of detection unit can there are many, one of embodiment is given below.
In one embodiment, detection unit includes state recognition module and protective cover board for contact rails defect state feature database,
In,
Protective cover board for contact rails defect state feature database, for storing the neural network model of protective cover board for contact rails;
State recognition module is connect for the image data of protective cover board for contact rails to be input in pixel peg model
The actual pixels data of the image of rail-engaging shield;By the image data of protective cover board for contact rails, actual pixels data and conductor rail
The location information of shield is input to the neural network model of protective cover board for contact rails, identifies the defect shape of protective cover board for contact rails
State information.
In the above-described embodiments, state recognition module may include expert's identifying system, for by protective cover board for contact rails
Image data is input in pixel peg model, obtains the actual pixels data of the image of protective cover board for contact rails;Conductor rail is prevented
The location information of the image data of shield, actual pixels data and protective cover board for contact rails, is input to the nerve of protective cover board for contact rails
Network model identifies the defect state information of protective cover board for contact rails.
Pixel peg model in the embodiment of the present invention can be indicated with following formula:
Wherein, s is longitudinal pixel;
D is the distance of default single picture;
R is the wheel circumference for runing vehicle;
Fn is the umber of pulse of trigger unit;
N is signal frequency split number;
Bn is signal frequency multiplication number.
By the way that the image data of protective cover board for contact rails is input to above-mentioned pixel peg model, available image is in reality
Pixel data in space.
Fig. 8 is the specific of the defect state information that state recognition module identifies protective cover board for contact rails in the embodiment of the present invention
Process, as shown in figure 8, the detailed process of the defect state information of identification protective cover board for contact rails includes:
Step 801, the image data of protective cover board for contact rails, the position letter of actual pixels data and protective cover board for contact rails are obtained
Breath;
Step 802, clear normal image data is picked out from the image data of protective cover board for contact rails, is proposed unintelligible
Abnormal image data;
Step 803, the relevant image data of protective cover board for contact rails is extracted from clear normal image data;
The relevant picture number of protective cover board for contact rails can be extracted by image processing algorithms such as edge extracting and projective transformations
According to;
Step 804, by the position of the relevant image data of protective cover board for contact rails, actual pixels data and protective cover board for contact rails
Information is input to the neural network model of protective cover board for contact rails, identifies the defect state information of protective cover board for contact rails;
The defect state information identified includes specific defect description, the actual pixels size of defect and the tool of defect
Body position (corresponds to the position on protective cover board for contact rails mileage).
Step 805, after testing staff checks the output information of the neural network model identification of protective cover board for contact rails
Form final testing result.
In one embodiment, the neural network model of protective cover board for contact rails is using sparse from coding learning algorithm building.
It is sparse from coding learning algorithm can the learning characteristic from no labeled data automatically, and provide more preferable than initial data
Feature description.
It is, of course, understood that the neural network model of protective cover board for contact rails can also be constructed using other methods, phase
Protection scope of the present invention should all be fallen by closing change case.
In one embodiment, using the mistake of the sparse neural network model from coding learning algorithm building protective cover board for contact rails
Journey is as follows:
Obtain the feature vector of the history image data of protective cover board for contact rails;
Utilize the neural network mould of the feature vector training protective cover board for contact rails of the history image data of protective cover board for contact rails
Type obtains the sparse features vector of the history image data of protective cover board for contact rails;
Using the sparse features vector sum feature vector of the history image data of protective cover board for contact rails, connecing after determining training
The neural network model of rail-engaging shield.
In the above-described embodiments, sparse from coding learning algorithm design neural network structure and definition cost function, and lead to
Cross one group of weight parameter w that iteration seeks making cost function to minimize (and corresponding intercept b), and obtain indicating characteristic
The sparse features vector σ (wx+b) of collection.
By taking the neural network model of the protective cover board for contact rails of 5 layers of structure as an example, comprising input layer, output layer, hidden layer L1,
L2 and Softmax layers of hidden layer, wherein input layer is image data layer, and 128 × 80 gray level image matrix transforms into one-dimensional vector
10240 nodes of input layer are connected afterwards;Hidden layer L1, hidden layer L2 constitute the depth network for indicating characteristics of image, set respectively
It is set to 256 and 196 nodes.It can be seen that original image is converted to lower through the sparse from after transcoding, coding transform of depth network
The expression feature of dimension.Feature vector after hidden layer L2 is transformed is connect with Softmax layers.
Hidden layer L1 and hidden layer L2 and input layer constitute a Bidirectional Conduction network, and its purpose is to establish reversed biography
Mechanism is led to seek the expression weight of sparse features, principle is to be defined as h by solving onew,b(x) function of ≈ x, so that
Sparse features vector σ (wx+b) expression approaches original image as far as possible, and correlation solves and derivation process is not reinflated herein.In defect
The identification link of state, such as shift defect identification, using the identification of defect state as a polytypic pattern recognition problem
It handles, Softmax layers are classified using Softmax regression model, which is a kind of machine learning side for having supervision
Method is popularization of the logistic regression model in more classification problems.
According to above-mentioned principle, Fig. 9 is that the neural network model of protective cover board for contact rails in the embodiment of the present invention implemented
Journey, as shown in figure 9, the process includes:
Step 901, the feature vector of the history image data of protective cover board for contact rails is obtained;
By taking the displacement of protective cover board for contact rails as an example, the history image data of protective cover board for contact rails can be the history figure of displacement
As data, history image data include positive sample and negative sample, positive sample, that is, protective cover board for contact rails normal condition data, negative sample
This is the disease state of protective cover board for contact rails, thus guarantee the robustness of the neural network model of protective cover board for contact rails, history figure
As the feature vector of data is displacement label vector y;
Step 902, the mind of the feature vector training protective cover board for contact rails of the history image data of protective cover board for contact rails is utilized
Through network model, the sparse features vector of the history image data of protective cover board for contact rails is obtained;
The history image data of protective cover board for contact rails are input to the neural network model of protective cover board for contact rails, and by repeatedly
Generation training obtains the characteristic parameter w and b of hidden layer L1 and hidden layer L2;
The history image data of protective cover board for contact rails are transformed to sparse features vector σ (wx+b);
Step 903, using the sparse features vector sum feature vector of the historical state data of protective cover board for contact rails, instruction is determined
The neural network model of protective cover board for contact rails after white silk;
By the feature vector of sparse features vector σ (wx+b) and history image data, (label vector y) is inputted together
Softmax layers, and repetitive exercise is completed, classifier is obtained, the neural network model of the protective cover board for contact rails after determining training.
The location information of the status data of protective cover board for contact rails and protective cover board for contact rails is input to and is connect by state recognition module
When the neural network model of rail-engaging shield, by the location information exhibition of the status data of protective cover board for contact rails and protective cover board for contact rails
At the input layer for the neural network model for connecting protective cover board for contact rails after one-dimensional vector, most afterwards through the nerve net of protective cover board for contact rails
The output layer of network model obtains output label, and classification corresponding to the output label is current protective cover board for contact rails defect state number
According to.
In one embodiment, the image data of protective cover board for contact rails includes:
File header information unit, the attribute information of the image data for storing protective cover board for contact rails;
An at least frame data unit, an at least frame image for the image data for storing protective cover board for contact rails.
Since the image storage data quantity of image acquisition units is very big, especially up to 4M, therefore high definition acquires the every frame of image
JPEG coded format can be selected to compress image, in big file memory module, image acquisition units are by every side contacts
The image of rail shield is stored as an independent monomer data file, and file size is according to acquisition mileage in approximate linear increasing
It is long, it is calculated according to 100Mb/ kilometers, at most can be extended to tens of G.In file format, File header information unit, for storing
The attribute information of the image data of protective cover board for contact rails, every frame image data are successively continuous as an independent frame data unit
It is stored in after header file information unit, a complete independent data files is formed, so as to store a large amount of image data.
Figure 10 is the form schematic diagram of the image data of protective cover board for contact rails in the embodiment of the present invention, as shown in Figure 10, conductor rail protection
Each image data file of the image data of cover includes a File header information unit and multiple frame data units.
Since protective cover board for contact rails is generally one group, in one embodiment, there are two data acquisition units, pacifies respectively
Loaded in operation vehicle car body both ends, and the level height of two data acquisition units difference within the set range.
Figure 11 is the system annexation figure in the embodiment of the present invention when there are two data acquisition unit, such as Figure 11 institute
Show, makes the difference of the level height of two data acquisition units not higher than 2mm by vehicle mounting process, to improve imaging precision.
In one embodiment, protective cover board for contact rails defect state information includes crack, crackle and the shifting of protective cover board for contact rails
Position information.
It is, of course, understood that protective cover board for contact rails defect state information can also include other kinds of disease, this
In repeat no more.
Protective cover board for contact rails condition detecting system proposed by the present invention can also include the function such as retrieval, inquiry, electronic report forms
Can, to realize the information management to image data and protective cover board for contact rails Disease Characters, the prior art is can be used in these functions
It realizes, the status data of protective cover board for contact rails, defect state information is possibly stored in database, to facilitate management.
In the protective cover board for contact rails condition detecting system that the embodiment of the present invention proposes, data acquisition unit is in operation vehicle fortune
When row, the status data of protective cover board for contact rails is acquired, it is anti-entire conductor rail can be acquired when conductor rail is to operation vehicle powered operation
The status data of shield;The positioning unit for being installed on operation vehicle vehicle bottom can run the position that protective cover board for contact rails is obtained when vehicle operation
Confidence breath;Finally, the location information of the detection unit status data of protective cover board for contact rails and protective cover board for contact rails based on the received,
The defect state information of protective cover board for contact rails can be directly obtained, is not necessarily to manual inspection, therefore detection speed is fast, measurement accuracy
It is high.
In addition, positioning unit can obtain accurate conductor rail according to its preset electronic labeling information in internal mileage position
The location information of shield, so that even if shield occurs movement and remains on the accurate position that can collect protective cover board for contact rails
Confidence breath, the precision of the defect state information of the protective cover board for contact rails obtained from are higher.Under the conditions of achievable non-uniform movement
Space equidistantly sample so that camera acquisition image more evenly, i.e., it is irredundant obtain protective cover board for contact rails image data.
Light Source Compensation module can obtain broader brightness range, so that camera has better image effect.Data acquisition unit is high
Integrated, miniaturization is spent, equipment volume is small, light-weight, the whole easy to use and flexible of rail-engaging shield condition detecting system, meets
To failure, direct integral replacing, plug and play.
The system that the embodiment of the present invention proposes is mainly used in urban track traffic electricity car, track inspection car, track work
Journey vehicle, track condition confirmation vehicle, urban rail synthetic detection vehicle etc., can be with rail detection system, rail polling system, rail wave
The organic integration such as Abrasion detecting system.
Based on same inventive concept, the embodiment of the present invention also proposes a kind of protective cover board for contact rails condition detection method, should
Method is applied to above-mentioned protective cover board for contact rails condition detecting system, and realization principle is similar with protective cover board for contact rails state-detection,
Which is not described herein again.
Figure 12 is the flow chart of protective cover board for contact rails condition detection method in the embodiment of the present invention, as shown in figure 12, the party
Method includes:
Step 1201, when runing vehicle operation, the status data of protective cover board for contact rails is acquired;
Step 1202, the location information that protective cover board for contact rails is obtained when vehicle operation is being runed;
Step 1203, it according to the location information of the status data of protective cover board for contact rails and protective cover board for contact rails, is contacted
The defect state information of rail shield.
In one embodiment, the status data of protective cover board for contact rails includes in image data, pressure data and temperature data
One of which or any combination.
In one embodiment, when runing vehicle operation, the status data of protective cover board for contact rails is acquired, comprising:
When runing vehicle operation, according to trigger signal, the image data of protective cover board for contact rails is acquired.
In one embodiment, when runing vehicle operation, according to trigger signal, the image data of protective cover board for contact rails is acquired,
Include:
When runing vehicle operation, according to the trigger signal of fixed frequency, protective cover board for contact rails is spatially equidistantly scanned,
Acquire the image data of protective cover board for contact rails.
In one embodiment, it according to the location information of the status data of protective cover board for contact rails and protective cover board for contact rails, obtains
The defect state information of protective cover board for contact rails, comprising:
The image data of protective cover board for contact rails is input in pixel peg model, the image of protective cover board for contact rails is obtained
Actual pixels data;
By the location information of the image data of protective cover board for contact rails, actual pixels data and protective cover board for contact rails, it is input to
The neural network model of protective cover board for contact rails identifies the defect state information of protective cover board for contact rails.
In one embodiment, the neural network model of protective cover board for contact rails is using sparse from coding learning algorithm building.
In one embodiment, using the mistake of the sparse neural network model from coding learning algorithm building protective cover board for contact rails
Journey is as follows:
Obtain the feature vector of the history image data of protective cover board for contact rails;
Utilize the neural network mould of the feature vector training protective cover board for contact rails of the history image data of protective cover board for contact rails
Type obtains the sparse features vector of the history image data of protective cover board for contact rails;
Using the sparse features vector sum feature vector of the history image data of protective cover board for contact rails, connecing after determining training
The neural network model of rail-engaging shield.
Based on above-mentioned multiple embodiments, the present invention proposes following one embodiment to illustrate protective cover board for contact rails state-detection
The detailed process of method, Figure 13 are the detail flowchart for the protective cover board for contact rails condition detection method that the embodiment of the present invention proposes,
As shown in figure 13, in one embodiment, the detailed process of protective cover board for contact rails condition detection method includes:
Step 1301, when runing vehicle operation, according to the trigger signal of fixed frequency, spatially equidistant scanning contact
Rail shield acquires the image data of protective cover board for contact rails;
Step 1302, the location information that protective cover board for contact rails is obtained when vehicle operation is being runed;
Step 1303, the image data of protective cover board for contact rails is input in pixel peg model, obtains conductor rail protection
The actual pixels data of the image of cover;
Step 1304, the position of the image data of protective cover board for contact rails, actual pixels data and protective cover board for contact rails is believed
Breath, is input to the neural network model of protective cover board for contact rails, identifies the defect state information of protective cover board for contact rails.
It is, of course, understood that the detailed process of above-mentioned protective cover board for contact rails condition detection method can also have other
Change case, associated change example should all fall into protection scope of the present invention.
In the protective cover board for contact rails condition detection method that the embodiment of the present invention proposes, it is applied to above-mentioned protective cover board for contact rails
Condition detecting system acquires the status data of protective cover board for contact rails when runing vehicle operation, can give operation vehicle power supply in conductor rail
The status data of entire protective cover board for contact rails is acquired when operation;It can believe runing the position for obtaining protective cover board for contact rails when vehicle operation
Breath;Finally, conductor rail can be directly obtained according to the location information of the status data of protective cover board for contact rails and protective cover board for contact rails
The defect state information of shield is not necessarily to manual inspection, therefore detection speed is fast, and measurement accuracy is high.
Furthermore it is possible to obtain accurate protective cover board for contact rails according to its preset electronic labeling information in internal mileage position
Location information, so that even if shield occurs movement and remains on the accurate location information that can collect protective cover board for contact rails,
The precision of the defect state information of the protective cover board for contact rails obtained from is higher.The space etc. under the conditions of non-uniform movement can be achieved
Spacing sampling, so that the image of camera acquisition is more evenly, i.e., the irredundant image data for obtaining protective cover board for contact rails.Light Source Compensation
Module can obtain broader brightness range, so that camera has better image effect.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the present invention, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
Particular embodiments described above has carried out further in detail the purpose of the present invention, technical scheme and beneficial effects
Describe in detail it is bright, it should be understood that the above is only a specific embodiment of the present invention, the guarantor being not intended to limit the present invention
Range is protected, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should be included in this
Within the protection scope of invention.
Claims (21)
1. a kind of protective cover board for contact rails condition detecting system characterized by comprising data acquisition unit, positioning unit and inspection
Survey unit, wherein
Data acquisition unit is installed on operation vehicle car body, is connect with detection unit, for acquiring contact when runing vehicle operation
The status data of rail shield, and the status data of protective cover board for contact rails is sent to detection unit;
Positioning unit is installed on operation vehicle vehicle bottom, connect with detection unit, for obtaining conductor rail protection when runing vehicle operation
The location information of protective cover board for contact rails is sent to detection unit by the location information of cover;
Detection unit is obtained for the status data of protective cover board for contact rails based on the received and the location information of protective cover board for contact rails
Obtain the defect state information of protective cover board for contact rails.
2. protective cover board for contact rails condition detecting system as described in claim 1, which is characterized in that the state of protective cover board for contact rails
Data include one of which or any combination in image data, pressure data and temperature data.
3. protective cover board for contact rails condition detecting system as claimed in claim 2, which is characterized in that further include: it is installed on operation
On vehicle wheel axle box, and the trigger unit being connect with data acquisition unit, for issuing trigger signal to data acquisition unit;
Data acquisition unit is specifically used for when runing vehicle operation, and trigger signal, acquires protective cover board for contact rails based on the received
Status data, and the image data of protective cover board for contact rails is sent to detection unit.
4. protective cover board for contact rails condition detecting system as claimed in claim 3, which is characterized in that data acquisition unit includes phase
Machine module, for scanning protective cover board for contact rails according to trigger signal, acquiring the image of protective cover board for contact rails when runing vehicle operation
Data.
5. protective cover board for contact rails condition detecting system as claimed in claim 4, which is characterized in that trigger unit is acquired to data
The trigger signal of unit sending fixed frequency;
Camera model is specifically used for: when runing vehicle operation, the trigger signal of fixed frequency based on the received, between spatially equal
Away from scanning protective cover board for contact rails, the image data of protective cover board for contact rails is acquired.
6. protective cover board for contact rails condition detecting system as described in claim 1, which is characterized in that data acquisition unit includes light
Source module and Light Source Compensation module, wherein
Light source module, for illuminating protective cover board for contact rails;
Light Source Compensation module, for highlighting to the protective cover board for contact rails that light source module illuminates.
7. protective cover board for contact rails condition detecting system as claimed in claim 2, which is characterized in that detection unit includes that state is known
Other module and protective cover board for contact rails defect state feature database, wherein
Protective cover board for contact rails defect state feature database, for storing the neural network model and pixel calibration mold of protective cover board for contact rails
Type;
State recognition module obtains conductor rail for the image data of protective cover board for contact rails to be input in pixel peg model
The actual pixels data of the image of shield;The image data of protective cover board for contact rails, actual pixels data and conductor rail are protected
The location information of cover is input to the neural network model of protective cover board for contact rails, identifies the defect state letter of protective cover board for contact rails
Breath.
8. protective cover board for contact rails condition detecting system as claimed in claim 7, which is characterized in that the nerve of protective cover board for contact rails
Network model is using sparse from coding learning algorithm building.
9. protective cover board for contact rails condition detecting system as claimed in claim 8, which is characterized in that using sparse from coding study
The process that algorithm constructs the neural network model of protective cover board for contact rails is as follows:
Obtain the feature vector of the history image data of protective cover board for contact rails;
Using the neural network model of the feature vector training protective cover board for contact rails of the history image data of protective cover board for contact rails, obtain
Obtain the sparse features vector of the history image data of protective cover board for contact rails;
Conductor rail using the sparse features vector sum feature vector of the history image data of protective cover board for contact rails, after determining training
The neural network model of shield.
10. protective cover board for contact rails condition detecting system as claimed in claim 2, which is characterized in that the figure of protective cover board for contact rails
As data include:
File header information unit, the attribute information of the image data for storing protective cover board for contact rails;
An at least frame data unit, an at least frame image for the image data for storing protective cover board for contact rails.
11. protective cover board for contact rails condition detecting system as described in claim 1, which is characterized in that data acquisition unit has two
It is a, it is respectively arranged in the both ends of operation vehicle car body, and the difference of the level height of two data acquisition units is within the set range.
12. protective cover board for contact rails condition detecting system as described in claim 1, which is characterized in that protective cover board for contact rails defect
Status information includes crack, crackle and the shift information of protective cover board for contact rails.
13. a kind of protective cover board for contact rails condition detection method, which is characterized in that it is any that this method is applied to claim 1 to 12
Protective cover board for contact rails condition detecting system described in, comprising:
When runing vehicle operation, the status data of protective cover board for contact rails is acquired;
Runing the location information that protective cover board for contact rails is obtained when vehicle operation;
According to the location information of the status data of protective cover board for contact rails and protective cover board for contact rails, the defect of protective cover board for contact rails is obtained
Status information.
14. protective cover board for contact rails condition detection method as claimed in claim 13, which is characterized in that the shape of protective cover board for contact rails
State data include one of which or any combination in image data, pressure data and temperature data.
15. protective cover board for contact rails condition detection method as claimed in claim 14, which is characterized in that when runing vehicle operation,
Acquire the status data of protective cover board for contact rails, comprising:
When runing vehicle operation, according to trigger signal, the image data of protective cover board for contact rails is acquired.
16. protective cover board for contact rails condition detection method as claimed in claim 15, which is characterized in that when runing vehicle operation,
According to trigger signal, the image data of protective cover board for contact rails is acquired, comprising:
When runing vehicle operation, according to the trigger signal of fixed frequency, protective cover board for contact rails, acquisition are spatially equidistantly scanned
The image data of protective cover board for contact rails.
17. protective cover board for contact rails condition detection method as claimed in claim 14, which is characterized in that according to protective cover board for contact rails
Status data and protective cover board for contact rails location information, obtain the defect state information of protective cover board for contact rails, comprising:
The image data of protective cover board for contact rails is input in pixel peg model, the reality of the image of protective cover board for contact rails is obtained
Pixel data;By the location information of the image data of protective cover board for contact rails, actual pixels data and protective cover board for contact rails, it is input to
The neural network model of protective cover board for contact rails identifies the defect state information of protective cover board for contact rails.
18. protective cover board for contact rails condition detection method as claimed in claim 17, which is characterized in that the mind of protective cover board for contact rails
Through network model using sparse from coding learning algorithm building.
19. protective cover board for contact rails condition detection method as claimed in claim 18, which is characterized in that using sparse from coding theory
The process for practising the neural network model of algorithm building protective cover board for contact rails is as follows:
Obtain the feature vector of the history image data of protective cover board for contact rails;
Using the neural network model of the feature vector training protective cover board for contact rails of the history image data of protective cover board for contact rails, obtain
Obtain the sparse features vector of the history image data of protective cover board for contact rails;
Conductor rail using the sparse features vector sum feature vector of the history image data of protective cover board for contact rails, after determining training
The neural network model of shield.
20. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor
Calculation machine program, which is characterized in that the processor realizes that claim 13 to 19 is any described when executing the computer program
Method.
21. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has perform claim
It is required that the computer program of 13 to 19 any the methods.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910375159.4A CN110044645B (en) | 2019-05-07 | Contact rail protective cover state detection system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910375159.4A CN110044645B (en) | 2019-05-07 | Contact rail protective cover state detection system and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110044645A true CN110044645A (en) | 2019-07-23 |
CN110044645B CN110044645B (en) | 2024-08-02 |
Family
ID=
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112208573A (en) * | 2020-09-17 | 2021-01-12 | 天津津航技术物理研究所 | Track defect detection system and method based on image recognition |
CN116001852A (en) * | 2023-03-24 | 2023-04-25 | 北京鼎昌复合材料有限责任公司 | Monitoring method and system for contact rail protection early warning |
Citations (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN201173768Y (en) * | 2008-04-08 | 2008-12-31 | 陈炳生 | Non-contact measurement device for measuring motion state red and hot metal outer diameter and size defect |
CN101351724A (en) * | 2005-11-02 | 2009-01-21 | 视声公司 | High frequency array ultrasound system |
CN102607977A (en) * | 2012-03-21 | 2012-07-25 | 西安交通大学 | Abrasion in-situ measuring device based on digital image processing and method |
CN102759347A (en) * | 2012-01-10 | 2012-10-31 | 广州科易光电技术有限公司 | Online in-process quality control device and method for high-speed rail contact networks and composed high-speed rail contact network detection system thereof |
CN104991502A (en) * | 2015-07-29 | 2015-10-21 | 成都国铁电气设备有限公司 | Vehicular overhead line system running state detection data analyzing and processing system and method |
CN105572543A (en) * | 2016-01-25 | 2016-05-11 | 成都国铁电气设备有限公司 | Overhead contact system operation state detecting device |
KR20160080884A (en) * | 2014-12-29 | 2016-07-08 | 한국철도기술연구원 | Contactless information transfer apparatus for railway vehicle |
CN206161036U (en) * | 2016-08-16 | 2017-05-10 | 北京华兴致远科技发展有限公司 | Pantograph and contact network state detection device |
CN107738665A (en) * | 2017-10-20 | 2018-02-27 | 珠海泰坦科技股份有限公司 | A kind of new conductor rail detection device |
CN108286944A (en) * | 2018-01-25 | 2018-07-17 | 中国铁道科学研究院 | Award streaming contact rail detection system in train lower part |
CN108594032A (en) * | 2018-01-05 | 2018-09-28 | 珠海泰坦科技股份有限公司 | A kind of contactless contact net on-line checking alarm real-time transmission system and method |
CN208278072U (en) * | 2018-06-04 | 2018-12-25 | 中铁武汉电气化局集团有限公司 | A kind of conductor rail detection vehicle |
CN208476808U (en) * | 2018-07-02 | 2019-02-05 | 常州沃翌智能科技有限公司 | Rail clip abnormality detection system |
CN109374054A (en) * | 2018-11-14 | 2019-02-22 | 湖南磁浮技术研究中心有限公司 | A kind of contact rail detection system and its application method |
KR20190024447A (en) * | 2017-08-31 | 2019-03-08 | 투아이시스(주) | Real-time line defect detection system |
CN209783926U (en) * | 2019-05-07 | 2019-12-13 | 中国铁道科学研究院集团有限公司 | state detection system for contact rail protective cover |
Patent Citations (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101351724A (en) * | 2005-11-02 | 2009-01-21 | 视声公司 | High frequency array ultrasound system |
CN201173768Y (en) * | 2008-04-08 | 2008-12-31 | 陈炳生 | Non-contact measurement device for measuring motion state red and hot metal outer diameter and size defect |
CN102759347A (en) * | 2012-01-10 | 2012-10-31 | 广州科易光电技术有限公司 | Online in-process quality control device and method for high-speed rail contact networks and composed high-speed rail contact network detection system thereof |
CN102607977A (en) * | 2012-03-21 | 2012-07-25 | 西安交通大学 | Abrasion in-situ measuring device based on digital image processing and method |
KR20160080884A (en) * | 2014-12-29 | 2016-07-08 | 한국철도기술연구원 | Contactless information transfer apparatus for railway vehicle |
CN104991502A (en) * | 2015-07-29 | 2015-10-21 | 成都国铁电气设备有限公司 | Vehicular overhead line system running state detection data analyzing and processing system and method |
CN105572543A (en) * | 2016-01-25 | 2016-05-11 | 成都国铁电气设备有限公司 | Overhead contact system operation state detecting device |
CN206161036U (en) * | 2016-08-16 | 2017-05-10 | 北京华兴致远科技发展有限公司 | Pantograph and contact network state detection device |
KR20190024447A (en) * | 2017-08-31 | 2019-03-08 | 투아이시스(주) | Real-time line defect detection system |
CN107738665A (en) * | 2017-10-20 | 2018-02-27 | 珠海泰坦科技股份有限公司 | A kind of new conductor rail detection device |
CN108594032A (en) * | 2018-01-05 | 2018-09-28 | 珠海泰坦科技股份有限公司 | A kind of contactless contact net on-line checking alarm real-time transmission system and method |
CN108286944A (en) * | 2018-01-25 | 2018-07-17 | 中国铁道科学研究院 | Award streaming contact rail detection system in train lower part |
CN208278072U (en) * | 2018-06-04 | 2018-12-25 | 中铁武汉电气化局集团有限公司 | A kind of conductor rail detection vehicle |
CN208476808U (en) * | 2018-07-02 | 2019-02-05 | 常州沃翌智能科技有限公司 | Rail clip abnormality detection system |
CN109374054A (en) * | 2018-11-14 | 2019-02-22 | 湖南磁浮技术研究中心有限公司 | A kind of contact rail detection system and its application method |
CN209783926U (en) * | 2019-05-07 | 2019-12-13 | 中国铁道科学研究院集团有限公司 | state detection system for contact rail protective cover |
Non-Patent Citations (1)
Title |
---|
戴鹏 等: "基于半监督深度学习的无砟轨道扣件缺陷图像识别方法", 《中国铁道科学》, vol. 39, no. 4, pages 43 - 48 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112208573A (en) * | 2020-09-17 | 2021-01-12 | 天津津航技术物理研究所 | Track defect detection system and method based on image recognition |
CN112208573B (en) * | 2020-09-17 | 2022-04-01 | 天津津航技术物理研究所 | Track defect detection system and method based on image recognition |
CN116001852A (en) * | 2023-03-24 | 2023-04-25 | 北京鼎昌复合材料有限责任公司 | Monitoring method and system for contact rail protection early warning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110660222B (en) | Intelligent environment-friendly electronic snapshot system for black-smoke road vehicle | |
CN102708378B (en) | Method for diagnosing fault of intelligent traffic capturing equipment based on image abnormal characteristic | |
CN205632521U (en) | Track state intelligence inspection device | |
CN112800911B (en) | Pavement damage rapid detection and natural data set construction method | |
US20130170701A1 (en) | Computer-readable recording medium and road surface survey device | |
CN102759347B (en) | Online in-process quality control device and method for high-speed rail contact networks and composed high-speed rail contact network detection system thereof | |
CN111855664A (en) | Adjustable three-dimensional tunnel defect detection system | |
CN103778681A (en) | Vehicle-mounted high-speed road inspection system and data acquisition and processing method | |
CN203246483U (en) | Intelligent detection system for moving state of locomotive walking part | |
CN105905132A (en) | Intelligent rail state patrolling device and method | |
CN103837087B (en) | Pantograph automatic testing method based on active shape model | |
JP3456339B2 (en) | Object observation method, object observation device using the method, traffic flow measurement device and parking lot observation device using the device | |
WO2023091447A1 (en) | Road surface conditions detection by distributed optic fiber system | |
CN106338335A (en) | Image-based sleeper vibration online monitoring method | |
CN116448773A (en) | Pavement disease detection method and system with image-vibration characteristics fused | |
CN112113649A (en) | Two-dimensional monitoring system and method for bridge structure response under vehicle load and action thereof | |
CN105674893B (en) | Absolute grating scale based on cmos image sensor and its measurement method | |
CN110044645A (en) | Protective cover board for contact rails condition detecting system and method | |
CN102622582B (en) | Road pedestrian event detection method based on video | |
CN115540770A (en) | Method for quickly and automatically measuring limit of trackside equipment | |
CN116740647A (en) | High-voltage transmission line galloping monitoring method based on binocular camera and deep learning | |
CN115760720A (en) | Crack online detection method and system based on mobile device and super-resolution reconstruction segmentation network | |
Wolf et al. | Asset Detection in Railroad Environments using Deep Learning-based Scanline Analysis. | |
CN202887450U (en) | Taxi anti-fake system | |
CN115169466A (en) | Method and device for drawing image of land, electronic equipment and computer readable medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant |