CN108128323A - With the selection method of the relevant laser image characteristic quantity of rail wear amount - Google Patents
With the selection method of the relevant laser image characteristic quantity of rail wear amount Download PDFInfo
- Publication number
- CN108128323A CN108128323A CN201711419564.9A CN201711419564A CN108128323A CN 108128323 A CN108128323 A CN 108128323A CN 201711419564 A CN201711419564 A CN 201711419564A CN 108128323 A CN108128323 A CN 108128323A
- Authority
- CN
- China
- Prior art keywords
- characteristic quantity
- rail
- laser
- laser image
- abrasion
- 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
- 238000010187 selection method Methods 0.000 title claims abstract description 5
- 238000005299 abrasion Methods 0.000 claims abstract description 70
- 238000006243 chemical reaction Methods 0.000 claims abstract description 4
- 238000001514 detection method Methods 0.000 abstract description 51
- 238000000034 method Methods 0.000 description 26
- 238000012360 testing method Methods 0.000 description 12
- 238000012545 processing Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 5
- 229910000831 Steel Inorganic materials 0.000 description 4
- 238000005452 bending Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 230000002596 correlated effect Effects 0.000 description 4
- 238000003708 edge detection Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 239000010959 steel Substances 0.000 description 4
- 239000012141 concentrate Substances 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000001914 filtration Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 2
- 230000003137 locomotive effect Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 235000006508 Nelumbo nucifera Nutrition 0.000 description 1
- 240000002853 Nelumbo nucifera Species 0.000 description 1
- 235000006510 Nelumbo pentapetala Nutrition 0.000 description 1
- 230000001154 acute effect Effects 0.000 description 1
- 210000001367 artery Anatomy 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000000205 computational method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000010415 tropism Effects 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
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
This divisional application discloses a kind of and relevant laser image characteristic quantity of rail wear amount selection method, belong to detection field, in combination of the selection for the laser image characteristic quantity of judgement, assuming that M is that collected feature samples set is pinpointed on the rail with abrasion, the set includes the laser image characteristic quantity of the reaction abrasion of N number of fixed point, correlation coefficient is selected as metric parameter, two laser images are characterized in relevant redundancy feature, only select one of laser image characteristic quantity judged as rail wear, to obtain effectively judging the combination of rail wear width and the minimum laser image characteristic quantity of depth, and a certain characteristic quantity of rail wear is calculated with the combination of this feature amount.
Description
The application is application number 201610765945.1, applying date 2016-08-30, denomination of invention " examine automatically by rail wear
The divisional application of survey device ".
Technical field
The invention belongs to detection fields, are related to a kind of rail wear automatic detection device, are more particularly to swashed based on a wordline
Light image processing and microprocessor can effectively to detect a kind of rail wear of rail head of rail surface abrasion depth and width automatic
Detection device.
Background technology
Railway is the main artery of communications and transportation, and compared to other means of transportation, heavy haul railway transport is big, at low cost with freight volume
Feature is developed rapidly all over the world.In track equipment, rail is most important building block, directly bears train load
Lotus simultaneously guides wheel to run.Whether the state of the art of rail is intact, and can directly affect train safe, flat by defined speed
Steady and continual operation.Railway locomotive is to transmit driving force and brake force by the frictional force between wheel track, and between wheel track
Friction can then lead to the generation of rail wear.With the high speed of locomotive, heavy duty, high density operation, the abrasion of rail will be quick
The increase of degree, particularly sharp radius curve outer rail medial surface abrasion are particularly acute.
The detection technique of rail wear have passed through monocular of conforming to the principle of simplicity measure ruler class tool detection, digitized instrument detection waited
Journey.At present, there are the main methods such as the measurement of contact fixture, EDDY CURRENT, optical triangulation in China in terms of Rail Abrasion Detection System,
The experience that testing result is often depending on the attitude of detection workman and instrument uses, these methods there is detection efficiency it is low, inspection
The not high problems of precision are surveyed, have been unable to meet the development need of current high speed.Although occur now a kind of using sharp
The detection device of light detection rail surface abrasion, but accurately step-by-step movement detection is can not achieve, other detection methods are also only stopped
Stay in theoretical research level.
Invention content
A kind of detection device that can perform Rail Abrasion Detection System method in order to obtain, the present invention propose a kind of rail mill
Detection device is consumed, to perform rail wear method, is characterized in that:Including a wordline laser device, ccd image sensor and micro-
Plane where the wordline light beam that processor, a wordline laser device are sent out and tested Rail Surface are in 60 ° of angles, ccd image sensing
The surface of plane where device is located at a wordline light beam, the input terminal of the microprocessor receive ccd image sensor acquisition
Rail image information, and perform Rail Abrasion Detection System method.
Advantageous effect:The present invention gives a kind of Rail Abrasion Detection System device, and laser and guide rail surface are angled
Irradiation, for ensure rail have the two sections of laser images formed during abrasion on surface perpendicular to the direction of guide rail not in straight line
On;60 ° are the right angled triangles that a special triangle, i.e., one 60 ° are formed on abrasion section for two sections of images, in this way
Facilitate calculating abrasion width and depth.
Description of the drawings
Fig. 1 is the structure diagram of rail wear automatic detection device described in embodiment 2;
Fig. 2 is without abrasion laser image schematic diagram;
Fig. 3 is has abrasion laser image schematic diagram;
Fig. 4 is brightness curve and disk diameter schematic diagram;
Fig. 5 is the label schematic diagram of data point and characteristic quantity.
Specific embodiment
Embodiment 1:A kind of rail wear automatic testing method, acquires the laser image of rail, and with complete rail laser
Light belt image carries out image comparison, to judge to detect whether rail has abrasion, judges that rail has abrasion, to the laser figure of acquisition
As carrying out laser image processing, the laser image processing includes image preprocessing and Edge extraction, laser image processing
Afterwards, select and extract with the relevant laser image characteristic quantity of rail wear amount, rail wear depth and width are calculated.Its
In:The extraction is one or more of following characteristics amount with the relevant laser image characteristic quantity of rail wear amount:
1) the length l of two sections of straight line portions of laser imageAAnd lB;
2) the width difference e of two sections of linear laser images;
3) the lengthwise position difference z of two sections of linear laser images;
4) between two sections of linear laser images changeover portion length lC;
5) between two sections of linear laser images changeover portion inclination angle theta;
Abrasion width and abrasion depth are referred to as the characteristic quantity of rail wear, select above-mentioned middle one or more laser images
Characteristic quantity, for calculating the depth and width of rail wear;
And in the depth and width for calculating rail wear, and non-selection whole above-mentioned laser image characteristic quantity is counted
It calculates, in order to optimize calculating process, selects base of the combination of laser image characteristic quantity as the depth and width for calculating rail wear
Plinth calculates data, and the method during combination is selected to be:The determining first and relevant laser image characteristic quantity of the wear characteristics, and from
Preferred features amount is selected in laser image characteristic quantity, calculates the degree of correlation system of remaining each laser image characteristic quantity and preferred features amount
Number, and be averaged, which is the threshold value beta of characteristic quantity selection, if wherein between certain two laser image feature
The absolute value of related coefficient | rTij | >=β, two laser images are characterized in relevant redundancy feature, only select it is one of as
The laser image characteristic quantity that rail wear judges.I.e. in combination of the selection for the laser image characteristic quantity of judgement, it is assumed that M is
Collected feature samples set is pinpointed on the rail with abrasion, the reaction abrasion which includes N number of fixed point swash
Light image characteristic quantity selects correlation coefficient as metric parameter, the parameter be characterized by between similitude, two laser figures
As being characterized in relevant redundancy feature, one of laser image characteristic quantity judged as rail wear is only selected, for finding
It can effectively judge the combination of rail wear width and the minimum laser image characteristic quantity of depth.
As a kind of embodiment, above-mentioned middle correlation coefficient is as metric parameter, to be characterized by the specific of a correlation
Method is:It is respectively using two groups of different laser image features are set:Ti={ tik, k=1,2 ..., n } and Tj={ tjk, k=1,
2 ..., n }, wherein k represents k-th of test point, shares n test point, then the related coefficient of two groups of laser image features defines such as
Under:
In formula,WithRespectively two groups of feature TiAnd TjAverage value:With
Correlation coefficient r TijReflect two groups of feature TiAnd TjDegree of correlation, rTijValue when being negative, represent two features
It is negatively correlated;rTijValue for timing, represent two feature positive correlations;Work as rTijBe when=0, between two wear characteristics it is incoherent,
Work as rTijAbsolute value closer to 1 when, the degree of correlation of two laser image features is higher, and the redundancy of generation is bigger, is reacting
In the laser image characteristic set of rail wear, using the related coefficient between each laser image characteristic quantity, threshold value beta is set, if
The absolute value of related coefficient wherein between certain two laser image feature | rTij| >=β, two laser images are characterized in correlation
Redundancy feature only selects one of laser image characteristic quantity judged as rail wear.
In another embodiment, it is for the determining method of threshold value beta described above:The laser image single to Mr. Yu is special
Sign amount chooses it to judge that remaining laser image characteristic quantity is for the method for the possibility of redundancy feature as preferred features:It determines
After preferred features, by calculating obtain in rail wear width and depth associated laser characteristics of image set with preferred features amount it
Between correlation coefficient, the mean value of this group of correlation coefficient data is set as threshold value beta, the determining method of threshold value beta is:
Wherein:The quantity of the c amounts of being characterized in formula, l are the serial number of preferred characteristic quantity, and j is the serial number of alternative features amount.
Above-described embodiment as a result, acquires the related coefficient between feature to obtain the possibility size of redundancy between feature, by this
The mean value of group correlation coefficient data is set as threshold value beta, using this threshold value as judging characteristic whether the foundation of redundancy, so as to sentence
During disconnected two feature redundancy, a laser image feature as the depth and width for calculating abrasion between redundancy feature is only selected,
To optimize calculating process, minimal features combination is obtained with this.
As a kind of embodiment, the computational methods of specific open abrasion depth and width:Two sections of straight line portions of laser image
The length l dividedAAs the preferred features of abrasion width detection, the lengthwise position difference z of two sections of linear laser images is deep as abrasion
Spend the preferred features of detection;Two straight line portions of laser image are calculated by the related coefficient of two groups of laser image features
Length lAAnd lBAs the characteristic quantity of abrasion width detection, the lengthwise position difference z of two sections of linear laser images is as abrasion depth inspection
The characteristic quantity of survey, abrasion width calculation formula are:
Wherein, l is the width without wearing away rail;
Wearing away depth calculation formula is:
V=ztan60 °
As a kind of embodiment, described image pretreatment includes the following steps:
First by image gray processing, the histogram of gray level image is drawn, gray scale is found out and concentrates range;
Then using following formula, grey level enhancement is carried out to gray level image, is more clear image;
Wherein:A, b is respectively the left and right boundary point of gray value integrated distribution in gray level image histogram, and x, y are represented respectively
Gray value before and after grey level enhancement.
As a kind of embodiment, the method for described image edge extracting includes the following steps:
Appoint and take a medium filtering brightness curve for being distributed pixel in the horizontal direction, in the curve peak-peak both sides point
Not Qu Chu the maximum continuity point of brightness step variation, take between the midpoint p of two groups of continuity points and q, p and q distance as detection
Template diameter;
If the brightness of image is f (i, j), a round s (c, r) is taken in picture field, and as detection template, wherein c is circle
The heart, coordinate are (ic,jc), r is radius;
The set of s (c, r) interior pixel is defined, and remembers the brightness of pixel in round s and is:
Make a small range movement in the horizontal direction of the detection template center of circle, it is bright to calculate each pixel in the detection template of each position
Degree and, should in the range of brightness and maximum template center location, the as bright wisp Pixel-level roof edge point, using most
Small square law fitting a straight line, the straight line are a wordline laser picture centre line, and the small range is that a left side is put centered on the center of circle
The image interval of right each 2 times of radiuses, to obtain the length l of the two of laser image sections of straight line portionsAAnd lB, two sections of linear laser figures
The length l of changeover portion as betweenC。
The rail laser image of the present embodiment acquisition is compared with complete image, to judge whether to acquire the steel in image
There is abrasion in rail, and when being judged as abrasion, further select characteristic quantity so that abrasion is calculated, and first qualitatively judge abrasion, then
Quantitative calculating wears away the thinking of depth and width and in calculating process, characteristic quantity is selected, to optimize abrasion depth
With the calculating process of width.
Embodiment 2:As the supplement of 1 technical solution of embodiment or as a kind of individual embodiment:Abrasion mainly go out
The head of present rail, abrasion include top surface abrasion and side wear, and when detection must detect the two numerical value simultaneously, come comprehensive
Close the wear intensity for judging rail.The present embodiment utilizes one word laser beam of high intensity narrow beam, laser and a wordline light beam institute
In plane and tested Rail Surface in 60 ° of angles, high-resolution Array CCD sensor is located at the surface shooting of laser image
Laser image.There is bending in the Rail Surface light beam image for there are abrasion, the position occurred by bending point and bending degree
Determine the width and depth of rail wear.
Rail wear automatic detection device includes:A wordline laser device, microprocessor, performs list at ccd image sensor
Member, display and acousto-optic warning unit and interface unit.Ccd image sensor acquires laser image, the image information obtained
It is transferred to microprocessor to be analyzed and processed, extracts image border and center and fitting a straight line, form complete rail laser
Light belt image outline converts image information into rail profile parameter, stores the characteristic quantity of rail profile, and join with complete rail
Number is compared, and judges rail with the presence or absence of abrasion.It does not wear away and continues subsequent point detection;There are abrasion, further determine that
Abrasion loss, the depth and width including abrasion.Execution unit receives the control signal of microprocessor, controls the traveling of detection device
Direction and speed, adjust the orientation of ccd image sensor, the output terminal of microprocessor respectively with LCD display and sound-light alarm
System connects, and LCD display is used to show current location and the wear intensity of rail, and acoustooptic alarm system is used to that rail to be prompted to work as
There are abrasion in front position, need to repair.For interface unit for exchanging information with host computer, host computer can be further to wearing away position
The further fine processing of image put determines accurately abrasion loss.
Image preprocessing is processing stage early period of laser image edge extracting, first by image gray processing, draws gray scale
The histogram of image finds out gray scale and concentrates range, and using formula (1), (wherein a, b are respectively gray value in gray level image histogram
The left and right boundary point of integrated distribution, x, y represent the gray value before and after grey level enhancement respectively) grey level enhancement is carried out to gray level image,
It is more clear image.
The edge detection of a wordline laser image uses " ridge-shaped " edge detection method.Based on single pixel brightness
Edge detection method noise resisting ability is poor, in order to reduce the interference of picture noise, pixel brightness each in a certain region and
As " ridge-shaped " edge distinguishing rule.Since circle has each to same tropism, do not influenced by ridge-shaped edge direction, therefore,
The present invention uses plate way " ridge-shaped " edge detection method.By the appropriate disk detection template of size in a wordline laser image
It is moved in a certain range of both sides, when the graded of the brightness sum of pixel each in template meets certain requirements, template
Central point is ridge-shaped marginal point.
Appoint and take a medium filtering brightness curve for being distributed pixel in the horizontal direction, in the curve peak-peak both sides point
Not Qu Chu the maximum continuity point of brightness step variation, take between the midpoint p of two groups of continuity points and q, p and q distance as detection
Template diameter, as shown in Figure 3.
If the brightness of image is f (i, j), a round s (c, r) is taken in picture field, and as detection template, wherein c is circle
The heart, coordinate are (ic,jc), r is radius.Define the set of s (c, r) interior pixel:
And remember the brightness of pixel in round s and be:
Make a small range movement in the horizontal direction of the detection template center of circle, it is bright to calculate each pixel in the detection template of each position
Degree and, should in the range of brightness and maximum template center location, the as bright wisp Pixel-level roof edge point.Using most
Small square law fitting a straight line, the straight line are a wordline laser picture centre line.The marginal point and the straight line of fitting detected is such as
Shown in Fig. 4.
Further extraction and rail wear amount associated laser image feature amount, selection method and threshold value including characteristic quantity
It determines.
The invention mainly relates to rail wear width and depth, steel can be determined by the bending degree of laser image
The width and depth of rail abrasion have following characteristic quantity to can be used for selecting:
1) the length l of two sections of straight line portions of laser imageAAnd lB;
2) the width difference e of two sections of linear laser images;
3) the lengthwise position difference z of two sections of linear laser images;
4) between two sections of linear laser images changeover portion length lC;
5) between two sections of linear laser images changeover portion inclination angle theta.
One or more features amount can be selected for judging the depth and width of rail wear, in selection for judgement
During the combination of characteristic quantity, it is desirable that different category features have marked difference, and redundancy feature interference is avoided to judge.Assuming that M be with
Collected feature samples set is pinpointed on the rail of abrasion, which includes the wear characteristics of n fixed point.Select the degree of correlation
Coefficient as metric parameter, the parameter can be characterized by between similitude, for find can effectively judge that rail wear is wide
The combination of the minimal features amount of degree and depth.If two groups of different wear characteristics are respectively:Ti={ tik, k=1,2 ..., n } and
Tj={ tjk, k=1,2 ..., n }, wherein k represents k-th of test point, shares the related coefficient of n test point, then two groups of features
It is defined as follows:
In formula,WithRespectively two groups of feature TiAnd TjAverage value:With
Correlation coefficient r TijReflect two groups of feature TiAnd TjDegree of correlation, rTijValue when being negative, represent two features
It is negatively correlated;Value is timing, represents two feature positive correlations.Work as rTijIt is incoherent between two features when=0.Then work as rTij
Absolute value closer to 1 when, represent two features degree of correlation it is higher, issuable redundancy is bigger at this time.
In the characteristic set of rail wear, using the related coefficient between each characteristic quantity, threshold value beta is set, if wherein certain
The absolute value of related coefficient between two features | rTij| >=β illustrates that the two are characterized in relevant redundancy feature, can only select
One of characteristic quantity judged as rail wear.
To Mr. Yu's single features, and the direct relation of rail wear depth and width is bigger, judgment method is simpler, uses
Higher in the feasibility for judging abrasion loss, selected possibility is bigger, and selected feature is as preferred features.Judge a certain
It is characterized as the possibility of redundancy feature, according to its correlation with preferred features, correlation is higher, then as relevant redundancy feature
Possibility it is bigger.After preferred features are determined, by calculating obtain in rail wear width and depth correlated characteristic set with
The mean value of this group of data is set as threshold value beta by the correlation coefficient between preferred features amount, as shown in formula (4):
After determining characteristic quantity, the rail wear amount of test position is calculated, is stored and is shown, had and start when transfiniting abrasion
Acoustic-optic alarm.
For interface unit for exchanging information with host computer, host computer can be further to abrasion position in off-line case
The further fine processing of image determines accurately abrasion loss.Due to the adoption of the above technical scheme, a kind of steel provided in this embodiment
Rail abrasion automatic detection device has such advantageous effect, due to the method using image procossing, in the control of microprocessor
Under, be detached from the control of PC machine, device can under the setting of operating personnel automatic running.Equipment has certain integrity and actual effect
Property, it is not only easy to operate, testing result is accurate, but also manufacturing cost is low convenient for the use of testing staff.
Embodiment 3:A kind of Rail Abrasion Detection System device, including a wordline laser device, ccd image sensor and microprocessor
Plane where the wordline light beam that device, a wordline laser device are sent out and tested Rail Surface are in 60 ° of angles, ccd image sensor position
The surface of plane where a wordline light beam, the input terminal of the microprocessor receive the rail of ccd image sensor acquisition
Image information, and Rail Abrasion Detection System method is performed, which can be the detection method in embodiment 1.Laser is with leading
The angled irradiation of track surface, to ensure that rail has the two sections of laser images formed during abrasion on surface perpendicular to guide rail
Direction is not point-blank;60 ° be in order to two sections of images abrasion section on formed a special triangle, i.e., one 60 °
Right angled triangle so that it is convenient to calculate abrasion width and depth.
As a kind of embodiment, detection device further includes LCD display, acoustooptic alarm system and interface unit, described micro-
The output terminal of processor is connect respectively with LCD display, acoustooptic alarm system, interface unit, the interface unit again with it is upper
Machine connects to exchange the information between microprocessor and host computer.
As a kind of embodiment, the detection device further includes execution unit, and execution unit receives the control of microprocessor
Signal, execution unit include the first servo motor and the second servo motor, and microprocessor connects the first servo motor to control steel
Rail Abrasion detecting system direction of travel and speed, microprocessor connect the second servo motor, the second servo motor connection ccd image
Sensor, the second servo motor is to control the orientation of ccd image sensor.Control signal is sent out by processor, control performs
The servo motor of unit presses the fixed velocity form set time, determines next test point, and pass through motor and finely tune ccd image
The angle of sensor ensures laser beam and Rail Surface into 60 ° of angles.
As a kind of embodiment:The microprocessor includes:
Judgment module is worn away, acquires the laser image of rail, and carry out image with complete rail laser light belt image and compare,
To judge to detect whether rail has abrasion;
Computing module is worn away, judges that rail there are abrasion, select and extracts laser image relevant with rail wear amount is special
Sign amount, rail wear depth and/or width is calculated.
As a kind of embodiment, the detection device further includes:
Characteristic quantity selecting module selects the described and relevant laser image characteristic quantity of rail wear amount as in following characteristics amount
More than one:
1) the length l of two sections of straight line portions of laser imageAAnd lB;
2) the width difference e of two sections of linear laser images;
3) the lengthwise position difference z of two sections of linear laser images;
4) between two sections of linear laser images changeover portion length lC;
5) between two sections of linear laser images changeover portion inclination angle theta;
The process of characteristic quantity selecting module selection characteristic quantity is:The characteristic quantity of rail wear includes abrasion width and mill
Depth is consumed, selects one or more laser image characteristic quantities, for calculating the depth of rail wear and/or width, used in selection
When the combination of the laser image characteristic quantity of judgement, it is assumed that M is that collected feature samples are pinpointed on the rail with abrasion
Set, the set include the laser image characteristic quantity of the reaction abrasion of N number of fixed point, correlation coefficient are selected to join as measurement
Number, two laser images are characterized in relevant redundancy feature, only select one of laser image judged as rail wear
Characteristic quantity, to obtain effectively judging the combination of rail wear width and the minimum laser image characteristic quantity of depth, and with this feature
The a certain characteristic quantity of rail wear is calculated in amount combination.
Two laser images of the judgement are characterized in that the method for relevant redundancy feature is:It is selected from laser image characteristic quantity
Preferred features amount, calculates the correlation coefficient of remaining each laser image characteristic quantity and preferred features amount, and is averaged, this is flat
Mean value is the threshold value beta of laser image characteristic quantity selection, if the related coefficient wherein between certain two laser image feature is absolute
Value | rTij | >=β, then two laser images be characterized in relevant redundancy feature, only select one of as rail wear judgement
Laser image characteristic quantity.
The method for calculating the correlation coefficient is:
If two groups of different laser image features are respectively:Ti={ tik, k=1,2 ..., n } and Tj={ tjk, k=1,
2 ..., n }, wherein k represents k-th of test point, shares n test point, then the related coefficient of two groups of laser image features defines such as
Under:
In formula,WithRespectively two groups of feature TiAnd TjAverage value:With
Correlation coefficient r TijReflect two groups of feature TiAnd TjDegree of correlation, rTijValue when being negative, represent two features
It is negatively correlated;rTijValue for timing, represent two feature positive correlations;Work as rTijBe when=0, between two wear characteristics it is incoherent,
Work as rTijAbsolute value closer to 1 when, the degree of correlation of two laser image features is higher, and the redundancy of generation is bigger;
Calculate threshold value beta method be:
Wherein:The quantity of the c amounts of being characterized in formula, l are the serial number of preferred characteristic quantity, and j is the serial number of alternative features amount.
The length l of two sections of straight line portions of laser imageAAs the preferred features of abrasion width detection, two sections of linear lasers
Preferred features of the lengthwise position difference z of image as abrasion depth detection;It is calculated by the related coefficient of two groups of laser image features
Obtain the length l of two straight line portions of laser imageAAnd lBAs the characteristic quantity of abrasion width detection, two sections of linear laser figures
Characteristic quantities of the lengthwise position difference z of picture as abrasion depth detection, abrasion width calculation formula are:
Wherein, l is the width without wearing away rail;
Wearing away depth calculation formula is:
V=ztan60 °.
The laser image of the rail of acquisition is being extracted with before the relevant laser image characteristic quantity of rail wear amount, having and swashing
The step of light image is handled, the laser image processing include image preprocessing and Edge extraction;
Described image pretreatment includes the following steps:
First by image gray processing, the histogram of gray level image is drawn, gray scale is found out and concentrates range;
Then using following formula, grey level enhancement is carried out to gray level image, is more clear image;
Wherein:A, b is respectively the left and right boundary point of gray value integrated distribution in gray level image histogram, and x, y are represented respectively
Gray value before and after grey level enhancement;
Using the method for described image edge extracting, include the following steps:
Appoint and take a medium filtering brightness curve for being distributed pixel in the horizontal direction, in the curve peak-peak both sides point
Not Qu Chu the maximum continuity point of brightness step variation, take between the midpoint p of two groups of continuity points and q, p and q distance as detection
Template diameter;
If the brightness of image is f (i, j), a round s (c, r) is taken in picture field, and as detection template, wherein c is circle
The heart, coordinate are (ic,jc), r is radius;
The set of s (c, r) interior pixel is defined, and remembers the brightness of pixel in round s and is:
Make a small range movement in the horizontal direction of the detection template center of circle, it is bright to calculate each pixel in the detection template of each position
Degree and, should in the range of brightness and maximum template center location, the as bright wisp Pixel-level roof edge point, using most
Small square law fitting a straight line, the straight line are a wordline laser picture centre line, and the small range is that a left side is put centered on the center of circle
The image interval of right each 2 times of radiuses, to obtain the length l of the two of laser image sections of straight line portionsAAnd lB, two sections of linear laser figures
The length l of changeover portion as betweenC。
The preferable specific embodiment of the above, only the invention, but the protection domain of the invention is not
This is confined to, in the technical scope that any one skilled in the art discloses in the invention, according to the present invention
The technical solution of creation and its inventive concept are subject to equivalent substitution or change, should all cover the invention protection domain it
It is interior.
Claims (1)
1. a kind of and relevant laser image characteristic quantity of rail wear amount selection method, it is characterised in that:The spy of rail wear
Sign amount includes abrasion width and abrasion depth, one or more laser image characteristic quantities is selected, for calculating the depth of rail wear
Degree and/or width, in combination of the selection for the laser image characteristic quantity of judgement, it is assumed that M is on the rail with abrasion
Collected feature samples set is pinpointed, which includes the laser image characteristic quantity of the reaction abrasion of N number of fixed point, select phase
Pass degree coefficient is characterized in relevant redundancy feature, only selects one of them as rail as metric parameter, two laser images
The laser image characteristic quantity judged is worn away, to obtain effectively judging rail wear width and the minimum laser image characteristic quantity of depth
Combination, and a certain characteristic quantity of rail wear is calculated with the combination of this feature amount.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711419564.9A CN108128323B (en) | 2016-08-30 | 2016-08-30 | The selection method of laser image characteristic quantity relevant to rail wear amount |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711419564.9A CN108128323B (en) | 2016-08-30 | 2016-08-30 | The selection method of laser image characteristic quantity relevant to rail wear amount |
CN201610765945.1A CN106274979B (en) | 2016-08-30 | 2016-08-30 | Rail wear automatic detection device |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610765945.1A Division CN106274979B (en) | 2016-08-30 | 2016-08-30 | Rail wear automatic detection device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108128323A true CN108128323A (en) | 2018-06-08 |
CN108128323B CN108128323B (en) | 2019-09-17 |
Family
ID=57675472
Family Applications (4)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711419564.9A Expired - Fee Related CN108128323B (en) | 2016-08-30 | 2016-08-30 | The selection method of laser image characteristic quantity relevant to rail wear amount |
CN201711422597.9A Expired - Fee Related CN108189859B (en) | 2016-08-30 | 2016-08-30 | Method for judging two laser image characteristics as related redundant characteristics |
CN201610765945.1A Expired - Fee Related CN106274979B (en) | 2016-08-30 | 2016-08-30 | Rail wear automatic detection device |
CN201711421370.2A Active CN108177660B (en) | 2016-08-30 | 2016-08-30 | Steel rail abrasion detection method with laser image processing step |
Family Applications After (3)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711422597.9A Expired - Fee Related CN108189859B (en) | 2016-08-30 | 2016-08-30 | Method for judging two laser image characteristics as related redundant characteristics |
CN201610765945.1A Expired - Fee Related CN106274979B (en) | 2016-08-30 | 2016-08-30 | Rail wear automatic detection device |
CN201711421370.2A Active CN108177660B (en) | 2016-08-30 | 2016-08-30 | Steel rail abrasion detection method with laser image processing step |
Country Status (1)
Country | Link |
---|---|
CN (4) | CN108128323B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108662983A (en) * | 2016-08-30 | 2018-10-16 | 大连民族大学 | The method that Rail Abrasion Detection System calculates correlation coefficient |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107194921B (en) * | 2017-05-19 | 2020-04-07 | 华夏高铁技术有限公司 | Automatic identification method and device for rail surface light band of steel rail |
CN111457835A (en) * | 2020-03-30 | 2020-07-28 | 合肥康普曼数字技术有限公司 | Threaded hole tooth socket detection method for battery tray |
CN113172551B (en) * | 2020-05-29 | 2022-10-14 | 浙江大学 | Quantitative measurement method for surface machining quality of steel rail |
CN111830049B (en) * | 2020-07-24 | 2023-09-22 | 深圳市道通科技股份有限公司 | Brake disc wear area identification method and wear identification system |
CN113077437B (en) * | 2021-03-31 | 2023-07-25 | 上海晨兴希姆通电子科技有限公司 | Workpiece quality detection method and system |
CN116047020B (en) * | 2023-04-03 | 2023-07-25 | 杭州电子科技大学 | Online repair performance testing device and testing method for steel rail surface damage |
CN117029696B (en) * | 2023-10-08 | 2024-01-02 | 天津津铁供电有限公司 | Abrasion detection method and detection equipment for rigid suspension contact net |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0611331A (en) * | 1991-12-26 | 1994-01-21 | Tokimec Inc | Instrument and method for measuring undulating wear of rail |
EP1415885A1 (en) * | 2002-10-29 | 2004-05-06 | Franz Plasser Bahnbaumaschinen-Industriegesellschaft m.b.H. | Method for the contactless measurement of a transverse profile or the distance between the rails of a track |
CN1776364A (en) * | 2005-11-22 | 2006-05-24 | 北京航空航天大学 | Steel rail near laser visual dynamic measuring device and method |
CN101144714A (en) * | 2007-10-26 | 2008-03-19 | 北京航空航天大学 | Steel rail wearing integrative parameter vehicle-mounted dynamic measuring device and method |
CN101576375A (en) * | 2009-05-21 | 2009-11-11 | 北京航空航天大学 | Fast processing method of laser vision image of steel rail wear |
CN102565068A (en) * | 2011-12-12 | 2012-07-11 | 关持循 | Steel rail damage automatic detection device |
CN202400107U (en) * | 2011-10-19 | 2012-08-29 | 北京鼎汉检测技术有限公司 | Detection device for dynamically detecting abrasion of lateral sides of steel railway rails |
CN102749061A (en) * | 2012-07-26 | 2012-10-24 | 上海工程技术大学 | Steel rail abrasion measuring method based on dynamic template |
CN103171589A (en) * | 2011-12-26 | 2013-06-26 | 同济大学 | Rail outline detection method based on abrasion value |
CN104501768A (en) * | 2014-08-14 | 2015-04-08 | 武汉武大卓越科技有限责任公司 | Rail rigidity measuring method based on machine vision |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4198164A (en) * | 1976-10-07 | 1980-04-15 | Ensco, Inc. | Proximity sensor and method and apparatus for continuously measuring rail gauge |
JP4302288B2 (en) * | 2000-04-19 | 2009-07-22 | 東京計器株式会社 | Portable rail cross-section measuring device |
ITVE20000023A1 (en) * | 2000-05-12 | 2001-11-12 | Tecnogamma S A S Di Zanin E & | LASER EQUIPMENT FOR THE CONTROL OF THE RAILWAYS OF A RAILWAY LINE. |
DE10326033B4 (en) * | 2003-06-10 | 2005-12-22 | Hema Electronic Gmbh | Method for adaptive error detection on an inhomogeneous surface |
CN101532827B (en) * | 2009-04-15 | 2010-12-01 | 北京航空航天大学 | Deviation correction method for measurement of rail wear based on laser vision |
CN102175953A (en) * | 2011-02-28 | 2011-09-07 | 长沙理工大学 | Discharge online monitoring and fault positioning method for power transmission line insulator |
CN104296682A (en) * | 2014-10-28 | 2015-01-21 | 电子科技大学 | Contour registration method used in train guide rail contour measurement based on machine vision |
CN104442917A (en) * | 2014-12-31 | 2015-03-25 | 北京二七轨道交通装备有限责任公司 | Track detection method |
CN104899888B (en) * | 2015-06-18 | 2017-10-24 | 大连理工大学 | A kind of image sub-pixel edge detection method based on Legendre squares |
-
2016
- 2016-08-30 CN CN201711419564.9A patent/CN108128323B/en not_active Expired - Fee Related
- 2016-08-30 CN CN201711422597.9A patent/CN108189859B/en not_active Expired - Fee Related
- 2016-08-30 CN CN201610765945.1A patent/CN106274979B/en not_active Expired - Fee Related
- 2016-08-30 CN CN201711421370.2A patent/CN108177660B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0611331A (en) * | 1991-12-26 | 1994-01-21 | Tokimec Inc | Instrument and method for measuring undulating wear of rail |
EP1415885A1 (en) * | 2002-10-29 | 2004-05-06 | Franz Plasser Bahnbaumaschinen-Industriegesellschaft m.b.H. | Method for the contactless measurement of a transverse profile or the distance between the rails of a track |
CN1776364A (en) * | 2005-11-22 | 2006-05-24 | 北京航空航天大学 | Steel rail near laser visual dynamic measuring device and method |
CN101144714A (en) * | 2007-10-26 | 2008-03-19 | 北京航空航天大学 | Steel rail wearing integrative parameter vehicle-mounted dynamic measuring device and method |
CN101576375A (en) * | 2009-05-21 | 2009-11-11 | 北京航空航天大学 | Fast processing method of laser vision image of steel rail wear |
CN202400107U (en) * | 2011-10-19 | 2012-08-29 | 北京鼎汉检测技术有限公司 | Detection device for dynamically detecting abrasion of lateral sides of steel railway rails |
CN102565068A (en) * | 2011-12-12 | 2012-07-11 | 关持循 | Steel rail damage automatic detection device |
CN103171589A (en) * | 2011-12-26 | 2013-06-26 | 同济大学 | Rail outline detection method based on abrasion value |
CN102749061A (en) * | 2012-07-26 | 2012-10-24 | 上海工程技术大学 | Steel rail abrasion measuring method based on dynamic template |
CN104501768A (en) * | 2014-08-14 | 2015-04-08 | 武汉武大卓越科技有限责任公司 | Rail rigidity measuring method based on machine vision |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108662983A (en) * | 2016-08-30 | 2018-10-16 | 大连民族大学 | The method that Rail Abrasion Detection System calculates correlation coefficient |
Also Published As
Publication number | Publication date |
---|---|
CN108128323B (en) | 2019-09-17 |
CN108189859A (en) | 2018-06-22 |
CN108189859B (en) | 2020-02-14 |
CN106274979B (en) | 2018-06-22 |
CN108177660B (en) | 2020-07-14 |
CN106274979A (en) | 2017-01-04 |
CN108177660A (en) | 2018-06-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106274979B (en) | Rail wear automatic detection device | |
CN106441107B (en) | Rail wear automatic testing method | |
CN109969736A (en) | A kind of large size carrier strip deviation fault intelligent detecting method | |
González et al. | Automatic traffic signs and panels inspection system using computer vision | |
CN107341802B (en) | Corner sub-pixel positioning method based on curvature and gray scale compounding | |
Wu et al. | Train rail defect classification detection and its parameters learning method | |
WO2023045299A1 (en) | Road surface technical condition detection method and device based on three-dimensional contour | |
Wang et al. | Intelligent prediction of wear location and mechanism using image identification based on improved Faster R-CNN model | |
CN103837087B (en) | Pantograph automatic testing method based on active shape model | |
CN108711148A (en) | A kind of wheel tyre defect intelligent detecting method based on deep learning | |
CN206019585U (en) | Rail Abrasion Detection System system | |
CN110288032A (en) | A kind of vehicle driving trace type detection method and device | |
CN101499168A (en) | Structured light strip center extraction method based on ridge line tracing and Hessian matrix | |
CN107462587A (en) | A kind of the precise vision detecting system and method for flexible IC substrates bump mark defect | |
Ye et al. | Visual high-precision detection method for tool damage based on visual feature migration and cutting edge reconstruction | |
CN114812403A (en) | Large-span steel structure hoisting deformation monitoring method based on unmanned aerial vehicle and machine vision | |
CN104316530A (en) | Part detection method and application | |
CN103063674B (en) | Detection method for copper grade of copper block, and detection system thereof | |
Yang et al. | A study on the roughness detection for machined surface covered with chips based on deep learning | |
Liu et al. | A binocular reconstruction method fused with Laplacian image information for pavement texture evaluation | |
CN117085969B (en) | Artificial intelligence industrial vision detection method, device, equipment and storage medium | |
CN112504184B (en) | Rapid online quality inspection system for three-dimensional size of steel plate | |
CN106157316A (en) | A kind of rail fastener location algorithm based on geometric match and device | |
Liu et al. | Surface splicing defect analysis and application of polarization maintaining fiber using graph cut with illumination priors | |
CN103604388A (en) | Method for rapidly and accurately positioning special markers in three-dimensional measuring of large-dimension steel plates |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190917 |