CN108725511B - Real-time position correction method for rail corrugation measuring point - Google Patents

Real-time position correction method for rail corrugation measuring point Download PDF

Info

Publication number
CN108725511B
CN108725511B CN201810573690.8A CN201810573690A CN108725511B CN 108725511 B CN108725511 B CN 108725511B CN 201810573690 A CN201810573690 A CN 201810573690A CN 108725511 B CN108725511 B CN 108725511B
Authority
CN
China
Prior art keywords
image
rail
measuring point
real
boundary line
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.)
Active
Application number
CN201810573690.8A
Other languages
Chinese (zh)
Other versions
CN108725511A (en
Inventor
刘宏立
郑倩
李艳福
陈亮
马子骥
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hunan University
Original Assignee
Hunan University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hunan University filed Critical Hunan University
Priority to CN201810573690.8A priority Critical patent/CN108725511B/en
Publication of CN108725511A publication Critical patent/CN108725511A/en
Application granted granted Critical
Publication of CN108725511B publication Critical patent/CN108725511B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • Quality & Reliability (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

The invention discloses a real-time position correction method for a rail corrugation measuring point, which comprises the steps of obtaining an image of a rail to be detected; carrying out image contrast enhancement; extracting a rail top boundary line of the steel rail; positioning a laser measuring point; removing perspective distortion of the image; and correcting the position of the laser measuring point in real time. The method for correcting the real-time position of the wavy grinding chord measuring point of the steel rail accurately monitors and corrects whether the chord measuring point is kept in the effective range of the middle part of the rail top in real time in the process of dynamically measuring the wavy grinding, thereby effectively ensuring the accuracy of wavy grinding measuring data and laying a solid technical foundation for high-precision dynamic detection of rail abrasion based on laser displacement.

Description

Real-time position correction method for rail corrugation measuring point
Technical Field
The invention particularly relates to a real-time position correction method for a rail corrugation measuring point.
Background
With the development of economic technology and the improvement of living standard of people, railways have entered into the production and life of people, and bring endless convenience to the production and life of people. Rails are an important component of railway systems. After the steel rail is put into use, a regular irregularity similar to a wave shape appears at certain parts of the top of the steel rail along the longitudinal surface of the steel rail along with the increase of operation time, and the irregularity is called wave-shaped abrasion of the steel rail, which is called wave-shaped abrasion of the steel rail, and is called wave-shaped abrasion of the steel rail for short.
At present, the rail corrugation measurement methods mainly include three methods: manual caliper method, inertial reference method and chord measuring method.
The manual caliper method uses the fluctuation change of a vernier caliper which is in contact with the surface of the track along the 1m ruler walking belt as the corrugation curve of the section. This method is inefficient and the position of the zero point of contact of the straightedge with the track surface, affected by rail distortion and surface damage, is usually not in the same horizontal plane, affecting the accuracy of the measurement.
The inertial reference method is usually used for high-speed rail inspection vehicles, and is characterized in that an accelerometer is mounted on a vehicle body, a photoelectric displacement meter is mounted on an axle box, and the displacement of the axle box relative to an acceleration fixed point is measured to be used as the corrugation of the point. The method can accurately describe the rail corrugation with the wavelength from 100mm to 50m, but the measurement accuracy is greatly influenced by the running speed and the irregularity of the wheel tread.
The chord measuring method utilizes the inherent transfer function relationship between the chord measuring value and the corrugation value constructed by a plurality of displacement sensors, and carries out secondary processing on the chord measuring value by designing a corresponding inverse filter, so that the output waveform approximates to the real appearance of the orbital corrugation. Compared with an inertia reference method, the measurement value is not influenced by the running speed of the vehicle body, and the method is widely applied to rail maintenance.
At present, the commonly used chord survey models on the railway line mainly include 3 types: two-point chord, three-point equal chord and three-point off-chord. Three-point deviational chord is the most widely applied.
According to the requirements of BS EN 15610 and 2009 and 13231-3-2012, the positions of the chord measuring points should be within +/-10 mm from the center line of the top of the steel rail. However, in the dynamic detection process, the measurement error of the current three-point off-chord measurement method is still large due to the following two factors:
(1) line bend
The minimum radius of the curved track on the track is about 150m and the length of the train car is about 20m, from which it is easy to deduce that the maximum lateral deviation in the middle of the car is 340mm, which is far beyond the valid range of ± 10 mm. At this time, 3 measurement points will move synchronously toward the rail top edge side.
(2) Snaking of vehicle body
When the wheel pair rolls forwards, the wheel pair is slightly shifted leftwards or rightwards due to the fact that the track is not smooth, and the like, so that contact points of the left wheel and the right wheel on the steel rail are changed. Because the tread of the wheel has a certain inclination, the rolling circles of the left wheel and the right wheel have different radiuses, the wheel with the large rolling radius has high speed, and the forward running speed of the wheel on the other side is correspondingly slower. Therefore, the direction of the axle can be changed, the wheels can move forwards for a moment to the left and a moment to the right according to certain amplitude, and the wheel pair moves forwards like a snake and crawls on the track, namely the snake moves. At this point, the line connecting the 3 measurement points will be inclined to the rail head centerline.
Disclosure of Invention
The invention aims to provide a real-time position correction method for a rail corrugation measuring point, which can monitor and correct the position of the measuring point in real time so as to ensure the reliable and accurate operation of a chord measuring method.
The invention provides a real-time position correction method for a rail corrugation measuring point, which comprises the following steps:
s1, acquiring an image of a steel rail to be detected;
s2, carrying out image contrast enhancement on the image of the steel rail to be detected obtained in the step S1;
s3, extracting a rail top boundary line of the steel rail according to the image obtained in the step S2;
s4, positioning a laser measuring point in a chord measuring method;
s5, removing perspective distortion of the image of the positioned image including the laser measuring point obtained in the step S4;
and S6, correcting the position of the laser measuring point in the chord measuring method in real time according to the image obtained in the step S5.
In step S2, the image contrast enhancement is performed on the image, specifically, the contrast enhancement is performed on the image by using a homomorphic sharpness algorithm.
The step S3 of extracting the rail top boundary line of the steel rail specifically includes the following steps:
A. positioning the boundary area;
B. b, performing boundary line fitting on the boundary area positioned in the step A;
C. and performing condition inspection on the image subjected to boundary line fitting.
The boundary area is positioned in the step A, specifically, the boundary area is positioned by adopting the following formula:
in the formula RwidthDenotes the width of the rectangle, RheightDenotes the height of the rectangle, IwidthRepresenting the width of the whole image, alpha being a predetermined scaling factor, NpixelIndicating 8 the number of pixels of the communication area.
And B, boundary line fitting, namely performing boundary line fitting on the boundary region in the image by using Hough transformation.
C, performing condition inspection on the image, namely confirming an upper boundary line and a lower boundary line according to the high-low position relation of the boundary line obtained in the step B; and then carrying out condition inspection on the boundary line obtained in the step B under the condition that the lower part of the upper boundary line is supposed to be a shadow area and the upper part of the lower boundary line is supposed to be a shadow area, thereby obtaining a real rail top boundary line.
Step S4, positioning the laser measurement point in the chord line measurement method, specifically positioning by using the following steps:
a. acquiring an image by adopting a color area-array camera, and simultaneously indicating a measuring point by adopting a laser displacement sensor of red laser;
b. and positioning the laser measuring point by adopting an RGB threshold value method.
And b, positioning the laser measuring point by adopting an RGB threshold method, specifically, positioning by adopting the following steps:
(1) given an image R of the original color railcBinary image R of the same sizebwAnd let R bebw＝0；
(2) According to a preset threshold range TrLine-by-line searching of original color rail imagesRcIf the image R iscI in the ith column ofsPixel point to itheThe RGB value of each pixel point is within a preset red threshold range TrIf so, the binary image R is processedbwI of (1)sTo the itheThe ith column pixel value of the row is all assigned to 1;
(3) performing opening operation on the binary image R obtained in the step (2) through a morphological regionbwProcessing is carried out, so that noise points are removed;
(4) performing morphological dilation on the binary image R obtained in the step (3)bwThe upper pixel regions are communicated into a whole;
(5) resetting the threshold range to TwFor binary image RbwAnd performing secondary verification on the measuring point in the area marked as 1, and determining the area meeting the requirement as the final laser measuring point, thereby completing the positioning of the laser measuring point.
The removing of the image perspective distortion in step S5 is specifically to remove the image perspective distortion by using 4 vertices of the boundary line as feature points and using the following equation:
wherein (u, v) is the pixel point of the original image, (x ', y') is the converted image pixel point, k11、k12、k13、k21、k22、k23、k31And k32Are all transform coefficients.
Step S6, which is to perform real-time correction on the position of the laser measurement point in the chord line measurement method, specifically, the real-time correction is performed by the following steps:
1) the lateral position deviation w of the measuring point is calculated by the following formuladev：
In the formula DpIs the distance between the measuring point on the image and the inner boundary of the rail head, DbIs the distance, w, between the inner and outer boundaries of the top of the rail on the imagerailThe width between the inner and outer boundaries of the real rail top;
2) if w is obtained in step 1)devIf the absolute value of the laser beam is larger than the preset threshold value, the position of the laser measuring point is adjusted in real time until wdevIs less than or equal to a predetermined threshold value.
The method for correcting the real-time position of the wavy grinding chord measuring point of the steel rail accurately monitors and corrects whether the chord measuring point is kept in the effective range of the middle part of the rail top in real time in the process of dynamically measuring the wavy grinding, thereby effectively ensuring the accuracy of wavy grinding measuring data and laying a solid technical foundation for high-precision dynamic detection of rail abrasion based on laser displacement.
Drawings
Fig. 1 is a schematic diagram of a hardware implementation of the method of the present invention.
FIG. 2 is a process flow diagram of the method of the present invention.
FIG. 3 is a schematic diagram of the effects of the method of the present invention before and after image contrast enhancement.
FIG. 4 is a schematic diagram of the boundary region positioning effect of the method of the present invention.
FIG. 5 is a schematic diagram of the boundary line fitting effect of the method of the present invention.
FIG. 6 is a schematic diagram of the final effect of the rail top boundary line of the steel rail of the method of the present invention.
FIG. 7 is a schematic diagram illustrating the positioning effect of the laser measuring points according to the method of the present invention.
FIG. 8 is a schematic representation of a typical rail in an embodiment of the method of the present invention.
FIG. 9 is a schematic diagram showing the distribution of the positions of 4 different chord measuring points in the embodiment of the method of the present invention.
FIG. 10 is a diagram illustrating a correction effect in an embodiment of the method of the present invention.
FIG. 11 is a schematic representation of the geometrical position of the 60kg/m rail head boundary in an embodiment of the method of the present invention.
Detailed Description
Fig. 1 is a schematic diagram of a hardware implementation of the method of the present invention: the invention provides a real-time position correction method for a rail corrugation measuring point, which is realized by adopting hardware shown in figure 1: as can be seen from fig. 1, the laser displacement sensor is used for indicating and measuring the chord measuring point, the LED light source is used for lighting, the area array camera is used for photographing the steel rail chord measuring point, and the photographed image is uploaded to the computer through the switch for processing; in addition, after the computer calculates the correction amount through a correction algorithm, the position of the laser displacement sensor is moved through the stepping motor, and therefore the string measuring point is corrected in real time.
The correction algorithm of the invention specifically comprises the following steps:
s1, acquiring an image of a steel rail to be detected;
s2, carrying out image contrast enhancement on the image of the steel rail to be detected obtained in the step S1;
to assist in extracting the top-of-rail boundary line, the image contrast enhancement algorithm should consider the following several issues:
firstly, the method comprises the following steps: differences in rail surface illumination and reflectivity can interfere with positioning of the rail top boundary line: for example, natural light, especially sunlight, can obscure rail images and reduce differences between rail top boundaries and other locations. On the other hand, since a bright track-top tread band reflects more light than other rough areas, the boundaries of the area may be mistaken for track-top boundaries;
secondly, the method comprises the following steps: the rail top boundary line contains the longitudinal linearity characteristic: essentially, the rail top boundary lines are two straight lines along the longitudinal direction of the rail, and the light intensity and the surface reflectivity of the rail are almost stable and constant on the straight lines. In the process of positioning the rail top boundary line, the characteristic can be fully utilized;
thirdly, the method comprises the following steps: synchronous existence of shadow area near boundary line: viewing fig. 11, since the inclination angle of the railhead 1:20 straight line is close to 90 ° and the width is only 1.1mm, the area is almost free from light reflection. At the tail end of the R13 area, because the inclination angle of the circular arc is increased sharply, the light reflectivity is far lower than other parts of the rail head, and a long and narrow shadow band is formed just inside the boundary line close to the top of the rail synchronously, so that the accuracy of the extracted boundary line can be verified;
therefore, contrast enhancement is carried out on the image by adopting a homomorphic sharpness enhancement algorithm; homomorphic refinement, also known as homomorphic filtering, is a method of compressing the dynamic range of an image and enhancing the contrast of the image in the frequency domain. It represents each image f (x, y) by the product of its illumination component i (x, y) and reflection component r (x, y) based on an illumination-reflectivity model of the image;
taking fig. 3 as an example, and fig. 3(a) as an example of an original rail image, before homomorphic filtering, the original rail image is blurred due to the influence of sunlight, and the corresponding shadow areas on both sides of the top border line of the rail in the binary image of fig. 3(b) cannot be clearly found; after homomorphic filtering, the effect of the illumination component is weakened, the image contrast in fig. 3(c) is significantly enhanced, and the shaded areas on both sides are clearly shown in fig. 3(d), which is very beneficial in the subsequent border check;
s3, extracting a rail top boundary line of the steel rail according to the image obtained in the step S2; the method specifically comprises the following steps of:
A. positioning the boundary area;
taking the rail gray image after contrast enhancement in fig. 3(c) as an example, the binary edge image extracted by directly using the canny operator is shown in fig. 4(a), and besides the rail top boundary line, a large number of other irrelevant lines exist in the image. In order to remove these extraneous lines, the longitudinal linearity characteristic contained in the boundary line is exploited here.
The smallest rectangle surrounding each 8-connected pixel region is shown in fig. 4 (b). The track longitudinal direction is used for representing the width direction, the track gauge direction is used for representing the height direction, and according to the longitudinal linear characteristic, the following empirical conditions are met in the geometric dimension of the minimum rectangle belonging to the boundary area of the track top:
in the formula RwidthDenotes the width of the rectangle, RheightDenotes the height of the rectangle, IwidthRepresenting the width of the entire image, alpha being a predetermined scaling factor, NpixelRepresents 8 the number of pixels of the communication area;
the results after treatment using the above formula are shown in FIG. 4 (c). Obviously, except the pixels in the boundary area, the rest irrelevant pixels are effectively removed, so that the potential boundary line is clearly shown, and the next step of straight line fitting is facilitated;
B. b, performing boundary line fitting on the boundary area positioned in the step A; specifically, boundary line fitting is carried out on a boundary region in an image by adopting Hough transformation; taking fig. 5 as an example, since one straight line may be divided into a plurality of adjacent pixel segments, the threshold interval at which two adjacent straight lines are merged is set to 300, the threshold length at which the merged straight line is retained is set to 30, and the extracted potential rail top boundary line (indicated by a green solid line) is depicted in fig. 5. Obviously, the number of potential boundary lines on the graph is more than 2, which means that some interference terms still exist;
C. performing condition inspection on the image subjected to boundary line fitting; firstly, confirming an upper boundary line and a lower boundary line according to the height position relation of the boundary lines obtained in the step B; secondly, performing condition inspection on the boundary line obtained in the step B under the condition that the lower part of the upper boundary line is a shadow area and the upper part of the lower boundary line is a shadow area, so as to obtain a real rail top boundary line; the final result is shown in fig. 6;
s4, positioning a laser measuring point in a chord measuring method; specifically, the following steps are adopted for positioning:
a. acquiring an image by adopting a color area-array camera, and simultaneously indicating a measuring point by adopting a laser displacement sensor of red laser;
the camera used in the method is a color area-array camera, and the color of the red laser measuring point of the sensor is obviously different from the colors of other areas on the image, so that the method can directly adopt an RGB threshold value method to realize the positioning of the laser measuring point;
however, in actual testing we found that some rust spots on the surface of the rail may also be mistaken for laser measurement points, an example is shown in fig. 7 (a). An enlarged view of the laser measurement spot is shown in fig. 7 (b). In contrast, the latter has an inner light core of bright white color except for the red area of the outer ring. By combining the two characteristics, the laser measuring point can be positioned more accurately;
b. positioning a laser measuring point by adopting an RGB threshold value method; specifically, the following steps are adopted for positioning:
(1) given an image R of the original color railcBinary image R of the same sizebwAnd let R bebw＝0；
(2) According to a preset threshold range Tr(threshold value range T)rCan be set to TrSearching original color rail image R column by column { R > 215, G < 185, B < 200})cIf the image R iscI in the ith column ofsPixel point to itheThe RGB value of each pixel point is within a preset red threshold range TrIf so, the binary image R is processedbwI of (1)sTo the itheThe ith column pixel value of the row is all assigned to 1;
(3) performing opening operation on the binary image R obtained in the step (2) through a morphological regionbwProcessing is carried out, so that noise points are removed;
(4) performing morphological dilation on the binary image R obtained in the step (3)bwThe upper pixel regions are communicated into a whole;
(5) for the binary image R obtained in the step (4)bwCounting the measurement points in (1); as shown in fig. 7, the measurement points obtained in graph (c) are greater than 3, and noise points are clearly present; therefore, the algorithm also needs to perform the following steps;
(6) resetting the threshold range to Tw(threshold value range T)wCan be set to TwFor a binary image R ═ R > 250, G > 240, B > 240}), for a binary image RbwPerforming secondary verification on the measuring point in the area marked as 1, and determining the area satisfying the range as a final laser measuring point (as shown in fig. 7 (d)), thereby completing the positioning of the laser measuring point;
s5, removing perspective distortion of the image of the positioned image including the laser measuring point obtained in the step S4; specifically, the method adopts 4 vertexes of the boundary line as characteristic points, and removes perspective distortion of the image by adopting the following formula:
wherein (u, v) is the pixel point of the original image, (x ', y') is the converted image pixel point, k11、k12、k13、k21、k22、k23、k31And k32Are all transform coefficients;
s6, according to the image obtained in the step S5, correcting the position of the laser measuring point in the chord measuring method in real time, specifically, correcting in real time by adopting the following steps:
1) the lateral position deviation w of the measuring point is calculated by the following formuladev：
In the formula DpIs the distance between the measuring point on the image and the inner boundary of the rail head, DbIs the distance, w, between the inner and outer boundaries of the top of the rail on the imagerailThe width between the inner and outer boundaries of the real rail top; if wdevIf the position is more than 0, the position of the measuring point needs to be adjusted towards the inner side of the track, otherwise, the position of the measuring point needs to be adjusted towards the outer side of the track;
2) if w is obtained in step 1)devIs greater than a preset threshold (the threshold can be determined according to the requirements of the string measuring point and the specific size of the steel rail), the position of the laser measuring point is adjusted in real time until wdevIs less than or equal to a predetermined threshold value. During specific adjustment, simple closed-loop control is adopted.
The process of the invention is further illustrated below with reference to one example:
static state experiment:
static validity testing is used primarily to verify the performance of the location monitoring subsystem. 600 steel rail images are collected on 3 typical steel rail forms such as a new rail, an active steel rail and a steel rail after milling, grinding and turning, wherein each form accounts for 200. These three forms cover almost all possible rail surface reflection characteristics, examples of which are shown in fig. 8(a) - (c) in sequence. Image size 1280 × 1024, resolution 1mm × 1 mm;
in addition, for each type, the string measuring points are distributed at 4 different transverse positions of the rail top for testing, and the specific position distribution is shown in FIG. 9;
the statistics of the positioning errors of the string measuring points distributed at 4 different positions on the 3 steel rail forms are shown in the following table 1:
table 13 transverse deviation calculation errors under distribution of 4 different chord measuring point positions of steel rail
In the table, MAE represents the root mean square error, RSME represents the mean absolute error;
wherein N represents the number of test samples, xiAndthe exact and estimated values of the ith term, respectively.
Dynamic experiment:
in order to verify the performance, a simulated dynamic test was carried out in a laboratory environment with the aid of two movable 60kg/m rails;
firstly, normally placing two steel rails according to a standard gauge, and acquiring a rail corrugation value as a true value through a trolley type detection platform;
then, the front and back track gauges of the two tracks are manually adjusted, the positions of the system sine correcting measuring points shown in the figure 1 are not used in the advancing process of the experiment platform, and the obtained corrugation value is used as a result before the system shown in the figure 1 is added;
finally, still on a track with a disordered track gauge, the positions of the string measurement points are monitored and corrected during the advance of the experimental platform using the system shown in fig. 1, and the obtained corrugation values are taken as a result of the addition of the system shown in fig. 1.
The track length is 6m, and the vehicle speed is about 4 km/h. Qualitative and quantitative comparisons of the corrugation measurements obtained by turning off/on the system as shown in fig. 1 with the actual values are shown in fig. 10 and table 2, respectively.
TABLE 2 comparison of the error of the corrugation measurement values obtained by switching off/on the auxiliary system with the real values
Closure system Opening system
MAE(mm) 0.1216 0.0374
RMSE(mm) 0.1489 0.0475
In the table, MAE represents the root mean square error, RSME represents the mean absolute error;
wherein N represents the number of test samples, xiAndthe exact and estimated values of the ith term, respectively.
As can be seen from Table 2 and FIG. 10, the corrugation measurement error introduced by the track gauge confusion can be effectively corrected by the system shown in FIG. 1, the two indexes are respectively reduced from 0.1216 and 0.1489mm before correction to 0.0374 and 0.0475mm, and the corrugation measurement result is more approximate to the true undulation form of the track.

Claims (8)

1. A real-time position correction method for a rail corrugation measuring point comprises the following steps:
s1, acquiring an image of a steel rail to be detected;
s2, carrying out image contrast enhancement on the image of the steel rail to be detected obtained in the step S1;
s3, extracting a rail top boundary line of the steel rail according to the image obtained in the step S2;
s4, positioning a laser measuring point in a chord measuring method; specifically, the following steps are adopted for positioning:
a. acquiring an image by adopting a color area-array camera, and simultaneously indicating a measuring point by adopting a laser displacement sensor of red laser;
b. positioning a laser measuring point by adopting an RGB threshold value method; specifically, the following steps are adopted for positioning:
(1) given an image R of the original color railcBinary image R of the same sizebwAnd let R bebw＝0；
(2) According to a preset threshold range TrSearching original color rail image R row by rowcIf the image R iscI in the ith column ofsPixel point to itheThe RGB value of each pixel point is within a preset red threshold range TrIf so, the binary image R is processedbwI of (1)sTo the itheThe ith column pixel value of the row is all assigned to 1;
(3) performing opening operation on the binary image R obtained in the step (2) through a morphological regionbwProcessing is carried out, so that noise points are removed;
(4) performing morphological dilation on the binary image R obtained in the step (3)bwThe upper pixel regions are communicated into a whole;
(5) resetting the threshold range to TwFor binary image RbwMiddle markPerforming secondary verification on the measuring point for the area 1, and considering the area meeting the requirement as a final laser measuring point, thereby completing the coarse positioning of the laser measuring point;
s5, removing perspective distortion of the image of the positioned image including the laser measuring point obtained in the step S4;
and S6, correcting the position of the laser measuring point in the chord measuring method in real time according to the image obtained in the step S5.
2. The method for correcting the real-time position of the grinding string measuring point of the rail wave as claimed in claim 1, wherein the step S2 is to enhance the image contrast, specifically to enhance the image contrast by using a homomorphic sharpness algorithm.
3. The method for correcting the real-time position of the wavy grinding string measuring point of the steel rail according to claim 1, wherein the step of extracting the top boundary line of the steel rail in the step S3 is to extract the top boundary line of the steel rail by the following steps:
A. positioning the boundary area;
B. b, performing boundary line fitting on the boundary area positioned in the step A;
C. and performing condition inspection on the image subjected to boundary line fitting.
4. The method for correcting the real-time position of the measuring point of the grinding string of the rail wave as claimed in claim 3, wherein the boundary area is positioned in the step A by adopting the following formula:
in the formula RwidthDenotes the width of the rectangle, RheightDenotes the height of the rectangle, IwidthRepresenting the width of the entire image, alpha being a predetermined scaling factor, NpixelIndicating 8 the number of pixels of the communication area.
5. The method for correcting the real-time positions of the grinding points of the steel rail waves according to claim 3, wherein the step B is performed by boundary line fitting, specifically by using Hough transformation to perform boundary line fitting on a boundary region in an image.
6. The method for correcting the real-time positions of the measuring points of the grinding string of the steel rail wave as claimed in claim 3, wherein the step C is used for carrying out condition inspection on the image, specifically, firstly, an upper boundary line and a lower boundary line are confirmed according to the height position relation of the boundary lines obtained in the step B; and then carrying out condition inspection on the boundary line obtained in the step B under the condition that the lower part of the upper boundary line is supposed to be a shadow area and the upper part of the lower boundary line is supposed to be a shadow area, thereby obtaining a real rail top boundary line.
7. The method for correcting the real-time position of the measuring point of the grinding string of the rail wave according to any one of claims 1 to 6, wherein the step S5 is performed to remove the perspective distortion of the image, specifically, the method uses 4 vertexes of the boundary line as the feature points, and the method uses the following formula to remove the perspective distortion of the image:
wherein (u, v) is the pixel point of the original image, (x ', y') is the converted image pixel point, k11、k12、k13、k21、k22、k23、k31And k32Are all transform coefficients.
8. The method for correcting the real-time position of the rail wave grinding string measuring point according to any one of claims 1 to 6, wherein the step S6 is to correct the position of the laser measuring point in the string measuring method in real time, and specifically comprises the following steps:
1) the lateral position deviation w of the measuring point is calculated by the following formuladev：
In the formula DpIs the distance between the measuring point on the image and the inner boundary of the rail head, DbIs the distance, w, between the inner and outer boundaries of the top of the rail on the imagerailThe width between the inner and outer boundaries of the real rail top;
2) if w is obtained in step 1)devIf the absolute value of the laser beam is larger than the preset threshold value, the position of the laser measuring point is adjusted in real time until wdevIs less than or equal to a predetermined threshold value.
CN201810573690.8A 2018-06-06 2018-06-06 Real-time position correction method for rail corrugation measuring point Active CN108725511B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810573690.8A CN108725511B (en) 2018-06-06 2018-06-06 Real-time position correction method for rail corrugation measuring point

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810573690.8A CN108725511B (en) 2018-06-06 2018-06-06 Real-time position correction method for rail corrugation measuring point

Publications (2)

Publication Number Publication Date
CN108725511A CN108725511A (en) 2018-11-02
CN108725511B true CN108725511B (en) 2019-12-20

Family

ID=63932148

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810573690.8A Active CN108725511B (en) 2018-06-06 2018-06-06 Real-time position correction method for rail corrugation measuring point

Country Status (1)

Country Link
CN (1) CN108725511B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109855562B (en) * 2019-01-28 2020-08-25 武汉武大卓越科技有限责任公司 Method and device for measuring rail abrasion
CN112033316B (en) * 2020-07-20 2022-07-22 深圳市埃伯瑞科技有限公司 Track waveform determining method and device, electronic equipment and storage medium
CN113011283B (en) * 2021-03-01 2022-04-29 浙江大学 Non-contact type rail sleeper relative displacement real-time measurement method based on video
CN114384078B (en) * 2021-12-31 2023-11-24 武汉大学 Method and equipment for detecting wave abrasion damage of steel rail in positioning mode
CN115307599B (en) * 2022-08-11 2023-09-22 济南致通华铁测量技术有限公司 Railway track smoothness adjusting method
CN115523869B (en) * 2022-11-02 2023-03-14 湖南科天健光电技术有限公司 Laser measurement system, corrugation measuring vehicle and operation train

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2165915A2 (en) * 2008-09-23 2010-03-24 VolkerRail Nederland BV Monitoring a turnout of a railway or tramway line
CN103171589A (en) * 2011-12-26 2013-06-26 同济大学 Rail outline detection method based on abrasion value
CN206019585U (en) * 2016-08-30 2017-03-15 大连民族大学 Rail Abrasion Detection System system
FR3057380A1 (en) * 2016-10-10 2018-04-13 Sncf Reseau METHOD AND SYSTEM FOR DETECTING REDUCED RAIL-WHEEL ADHERENCE, AND VEHICLE EQUIPPED WITH SUCH A SYSTEM

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102014201729A1 (en) * 2014-01-31 2015-08-06 Siemens Aktiengesellschaft track vehicle

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2165915A2 (en) * 2008-09-23 2010-03-24 VolkerRail Nederland BV Monitoring a turnout of a railway or tramway line
CN103171589A (en) * 2011-12-26 2013-06-26 同济大学 Rail outline detection method based on abrasion value
CN206019585U (en) * 2016-08-30 2017-03-15 大连民族大学 Rail Abrasion Detection System system
FR3057380A1 (en) * 2016-10-10 2018-04-13 Sncf Reseau METHOD AND SYSTEM FOR DETECTING REDUCED RAIL-WHEEL ADHERENCE, AND VEHICLE EQUIPPED WITH SUCH A SYSTEM

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
地铁轨道波浪形磨耗检测系统研究;李宏锋;《中国优秀硕士论文全文数据库 工程科技Ⅱ辑》;20140915;第4章 *

Also Published As

Publication number Publication date
CN108725511A (en) 2018-11-02

Similar Documents

Publication Publication Date Title
CN108725511B (en) Real-time position correction method for rail corrugation measuring point
US6909514B2 (en) Wheel profile inspection apparatus and method
US11820178B2 (en) Tread depth measurement
CN107678036B (en) Vehicle-mounted non-contact type contact net geometric parameter dynamic detection system and method
RU2743020C2 (en) System and method for monitoring railway wheels
CN108778889B (en) Method and measuring system for detecting a fixed point beside a track
CN109855562B (en) Method and device for measuring rail abrasion
CN109017867B (en) Dynamic measuring method for rail corrugation
US8078025B2 (en) Vehicle dynamic measurement device and method for comprehensive parameters of rail wear
CN112116619B (en) Multi-line structured light system stripe center line extraction method based on structural constraint
CN104236498B (en) Method for measuring end size of switch rail
CN109653045B (en) Track gauge measuring method and device
DE102010002258A1 (en) Method and device for determining distances on a vehicle
Xing et al. Online detection system for wheel-set size of rail vehicle based on 2D laser displacement sensors
CN106996748A (en) A kind of wheel footpath measuring method based on binocular vision
CN105004280A (en) Image restoring method in train guiderail contour measurement based on machine vision
CN108917621B (en) Pantograph slide plate upper edge single-pixel tracking detection method
CN111968079B (en) Three-dimensional pavement crack extraction method based on local extremum of section and segmentation sparsity
CN110954005B (en) Medium-low speed maglev train suspension gap detection method based on image processing
CN109489623A (en) A method of using three-dimensional laser scanner measurement bridge approach differential settlement
CN114240916A (en) Multi-polarized light point cloud data fusion method and device for appearance state of steel rail
CN110458785B (en) Magnetic levitation ball levitation gap detection method based on image sensing
CN112964195B (en) Power supply rail geometric parameter comprehensive detection method and system based on laser triangulation method
CN115219251A (en) Railway vehicle dynamic performance monitoring system based on wheel pair operation posture
KR101371531B1 (en) A system and a method for inspecting catenary through a vibration correction

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
CB03 Change of inventor or designer information

Inventor after: Liu Hongli

Inventor after: Zheng Qian

Inventor after: Li Yanfu

Inventor after: Chen Liang

Inventor after: Ma Ziji

Inventor before: Liu Hongli

Inventor before: Li Yanfu

Inventor before: Chen Liang

Inventor before: Ma Ziji

CB03 Change of inventor or designer information
GR01 Patent grant
GR01 Patent grant