CN108917621B - Pantograph slide plate upper edge single-pixel tracking detection method - Google Patents
Pantograph slide plate upper edge single-pixel tracking detection method Download PDFInfo
- Publication number
- CN108917621B CN108917621B CN201810699726.7A CN201810699726A CN108917621B CN 108917621 B CN108917621 B CN 108917621B CN 201810699726 A CN201810699726 A CN 201810699726A CN 108917621 B CN108917621 B CN 108917621B
- Authority
- CN
- China
- Prior art keywords
- pantograph
- image
- pixel
- edge
- point
- 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
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 38
- 238000004364 calculation method Methods 0.000 claims abstract description 14
- 230000009466 transformation Effects 0.000 claims abstract description 13
- 238000003708 edge detection Methods 0.000 claims abstract description 10
- 238000000034 method Methods 0.000 claims description 12
- 238000005299 abrasion Methods 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000002093 peripheral effect Effects 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 5
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 9
- 230000003137 locomotive effect Effects 0.000 description 7
- 238000005457 optimization Methods 0.000 description 5
- 238000000605 extraction Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 230000004807 localization Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000011426 transformation method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
- G01B11/06—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness for measuring thickness ; e.g. of sheet material
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a pantograph slide plate upper edge single-pixel tracking detection method which comprises the steps of firstly, geometrically correcting a pantograph image through perspective transformation, then, searching an area where a pantograph slide plate is located in the image in an integral graph calculation mode, then, carrying out Canny edge detection on the pantograph image, positioning a contact net and the lower edge of the pantograph slide plate through Hough transformation on the basis, and finally, positioning the upper edge of the pantograph slide plate through a single-pixel tracking detection method. The technology can efficiently obtain the effective information of the pantograph slide plate, improves the accuracy of the edge information of the pantograph slide plate, and has strong practicability and wide application prospect.
Description
Technical Field
The invention relates to the technical field of on-line automatic detection of rail vehicles, in particular to the field of detection of pantograph abrasion values of rolling stock such as railways, urban rails, subways and the like, and specifically relates to a single-pixel tracking detection method for the upper edge of a pantograph slide plate.
Background
The pantograph is an important device for removing electricity from a power grid of the locomotive, and the running safety of the locomotive can be directly influenced if the pantograph can work normally. The abrasion of the pantograph pan is an important factor influencing the normal operation of the pantograph. Therefore, the abrasion value of the pantograph slide plate is accurately detected, and the method has important reference significance for guiding the crew in the engineering section to carry out overhead maintenance.
In the operation process of the electric locomotive, the electric locomotive needs to get electricity from a power grid through the contact between the pantograph slide plate and a contact network. This contact-type power supply inevitably leads to electrical wear and sliding wear of the pantograph pan. If the car is left to wear and not take action, the car will pull or even block the net, and even become the direct cause of serious railway traffic accident.
In the traditional method, the abrasion value of the pantograph slide plate is detected by manual measurement, parking detection is required, time and labor are wasted, and manual measurement errors exist. With the development of image processing technology, the technology of detecting the size, the shape and the like of an object through an acquired high-resolution image is developed and widely applied. The image detection technology includes: non-contact, detect accurate, detection efficiency height advantage. In image detection, whether the upper edge of the pantograph pan can be accurately detected or not becomes an important influence factor influencing the detection precision of the wear value of the pantograph pan. Therefore, the research on the detection method of the upper edge of the pantograph with high precision and high reliability has great significance for improving the detection level of the abrasion of the pantograph sliding plate of the locomotive.
Disclosure of Invention
The invention aims to solve the technical problem that the upper edge of the existing pantograph slide plate is not positioned in the aspect of detecting the abrasion value, and provides a single-pixel tracking detection method for the upper edge of the pantograph slide plate by utilizing an image processing technology, so that the upper edge of the pantograph slide plate can be positioned efficiently and accurately.
The invention adopts the following technical scheme for solving the technical problems:
a pantograph slide plate upper edge single-pixel tracking detection method comprises the following steps:
step 1: correcting trapezoidal distortion by applying a geometric correction technology to the pantograph image;
step 2: adopting an integral diagram calculation mode to obtain the area of the pantograph slide plate in the image;
and step 3: carrying out Canny edge detection on the area where the sliding plate is located to obtain an edge image of the pantograph sliding plate;
and 4, step 4: positioning the lower edges of the contact net and the pantograph slide plate by using a Hough detection technology;
and 5: and traversing upwards from the tail end of the lower edge of the pantograph to find the tail end of the upper edge, and performing single-pixel tracking towards the direction of a contact net to realize complete detection of the upper edge of the pantograph.
As a further optimization scheme of the pantograph pan upper edge single-pixel tracking detection method, the detailed steps of geometrically correcting the pantograph image in step 1 are as follows:
step 1.1: when the pantograph slide plate does not pass through, vertically placing a calibration checkerboard at the position of the photoelectric sensor, triggering an industrial camera, and acquiring a checkerboard image;
step 1.2: and manually selecting four vertexes of the 4 x 4 checkerboard area in the checkerboard image, acquiring the relation between the pixel value and the actual distance by taking the vertical direction as reference due to trapezoidal distortion, mapping the four vertexes into the four vertexes of the square area, and obtaining the four calculated vertexes as target vertexes. And obtaining a mapping transformation matrix through calculation.
Step 1.3: and applying the transformation matrix to the pantograph image to obtain the pantograph image without geometric distortion.
As a further optimization scheme of the pantograph pan upper edge single-pixel tracking detection method, the detailed step of performing pan area positioning on the pantograph image in step 2 is as follows:
step 2.1: 2448 and 300, defining a strip sliding window, traversing from the top of the image, and having a step size of 50;
step 2.2: and calculating the sum of pixels in the sliding window by adopting an integral graph mode, wherein the area where the pixel and the minimum sliding window are located is the area where the target sliding plate is located.
As a further optimization scheme of the pantograph slide plate upper edge single-pixel tracking detection method, the Canny edge detection in the step 3 comprises the following detailed steps:
step 3.1: the image is convolved in the x and y directions with the following convolution operators to compute the image gradient values:
two partial derivative matrices of the image in the x and y directions can be obtained by using a first order finite difference approximation to calculate the gradient. Due to the gradient vector, the magnitude and angle of the gradient are calculated as follows:
P[i,j]＝(f[i,j+1]-f[i,j]+f[i+1,j+1]-f[i+1,j])/2
Q[i,j]＝(f[i,j]-f[i+1,j]+f[i,j+1]-f[i+1,j+1])/2
θ[i,j]＝arctan(Q[i,j]/P[i,j])
where M [ i, j ] is the magnitude of the gradient and θ [ i, j ] is the direction of the gradient.
Step 3.2: after the gradient amplitude and the angle of the point C are calculated, in order to determine whether the point C is an edge of the image, the pixel value of the point C is compared with the pixel values in the 8 neighborhoods of the point C, if the gray value of the point C is not the maximum value, the point C is not the edge point of the image, and the gray value of the point C can be assigned to be 0. After the step is finished, a binary image containing image edge information is obtained, non-edge points are set to be 0, and gray values of the points which are possibly edges can be assigned to be 255;
step 3.3: finding out the maximum pixel value Max and the minimum pixel value Min of the image, and defining a high threshold value as TH0.7Max, low threshold TLIf the gradient value of the pixel point is higher than the high threshold, the point is determined to be an edge point, if the gradient value of the pixel point is lower than the low threshold, the point is determined to be a non-edge point, and if the gradient value of the pixel point is between the high threshold and the low threshold. And further inspecting the gradient value of the pixel point in the neighborhood of the pixel point 8, if the gradient value of the peripheral pixel point is higher than the high threshold value, determining the pixel point as an edge point, and if the gradient values of the peripheral pixel points are all smaller than the high threshold value, determining the pixel point as a non-edge point.
As a further optimization scheme of the pantograph slide plate upper edge single-pixel tracking detection method, the detailed steps of the Hough detection technology in the step 4 are as follows:
step 4.1: straight lines in the image are detected using the Hough transform line detection technique.
Step 4.2: and if the slope k of the straight line is less than 0.12 (the value of the slope is related to the shooting angle and the shooting position of the camera, 0.12 is the optimal value under the specific shooting condition), and the coordinates of the bottom of the target sliding plate area determined in the approximation step II are within the range of +/-50 pixels, the straight line is considered as the lower edge of the pantograph sliding plate, and if the k is more than 0.2 and less than 0.4, the straight line is considered as the pantograph contact network.
As a further optimization scheme of the pantograph slide plate upper edge single-pixel tracking detection method, an upper edge tracking algorithm in step 5 comprises the following detailed steps:
step 5.1: the end of the lower edge of the pantograph slide positioned above is first found. Because the contact abrasion of the pantograph slide plate is usually concentrated in the center of the pantograph slide plate, the tail end of the pantograph slide plate is usually the best state that the pantograph slide plate is not abraded, taking the right half-slide plate of the pantograph as an example, the first pixel point with the pixel value of 255 is searched upwards from the tail end of the lower edge of the right lower slide plate, namely the tail end position (i, j) of the upper edge of the pantograph slide plate.
Step 5.2: and in view of the fact that only pixel points with pixel values of only 0 and 255 are left in the pantograph image after the edge detection, searching from the upper edge end of the pantograph to the pantograph overhead line system. Searching positions (i-1, j-1), (i-1, j) and (i-1, j +1) by the right pantograph half skateboard, searching positions (i +1, j-1), (i +1, j) and (i +1, j +1) by the left pantograph half skateboard, wherein due to the continuity of the edge of the pantograph slide board, pixel points with the pixel value of 255 are necessary in the three positions, continuously searching downwards until the pixel value is 0, stopping searching downwards, and determining the position with the last pixel value of 255 as the upper edge of the pantograph slide board.
Step 5.3: and repeating the steps until the position of the overhead line system is reached.
Compared with the prior art, the invention adopting the technical scheme has the following technical effects:
the invention provides a pantograph slide plate upper edge single-pixel tracking detection method which comprises the steps of firstly carrying out geometric correction on a pantograph image, then utilizing an integral graph calculation mode to locate an area where a target slide plate is located, then carrying out edge extraction on the pantograph slide plate image by using Canny edge detection, locating a lower edge of the slide plate and a contact net by using a Hough line detection technology on the basis, and finally adopting an upper edge tracking algorithm to locate the upper edge of the pantograph slide plate.
Drawings
FIG. 1 is a schematic diagram of an image acquisition system layout;
FIG. 2 is a schematic flow chart of a pantograph pan upper edge single-pixel tracking detection method according to some embodiments of the present invention;
fig. 3(a) and 3(b) are schematic diagrams of a first original captured image and a second original captured image, respectively;
fig. 4(a), fig. 4(b), fig. 4(c), fig. 4(d), fig. 4(e), fig. 4(f) are the first checkerboard image, the second checkerboard image, the first geometry corrected checkerboard image, the second geometry corrected checkerboard image, the first original captured image corrected image, and the second original captured image corrected image, respectively;
FIGS. 5(a), 5(b) are first and second corrected sled area positioning images, respectively;
FIGS. 6(a), 6(b) are first and second corrected sled edge extraction images, respectively;
7(a) and 7(b) are respectively the first corrected slide plate lower edge and catenary positioning image and the second corrected slide plate lower edge and catenary positioning image;
FIGS. 8(a), 8(b) are a top edge positioning image after a first corrected sled has used single-pixel tracking and a top edge positioning image after a second corrected sled has used single-pixel tracking, respectively;
FIG. 9 is a schematic diagram of the manner of integral map calculation;
FIG. 10 is a schematic view of a right sled tracking orientation.
Detailed Description
The technical scheme of the invention is further explained in detail by combining the attached drawings:
fig. 1 is a schematic layout diagram of an image acquisition system used in this embodiment, which is an example of an exemplary layout manner of a locomotive pantograph wear detection system, where the image acquisition system is used to acquire a real-time image of a locomotive pantograph and includes an image acquisition camera, an infrared trigger switch, a light source, and the like, which are arranged above a track.
Fig. 2 is a schematic flow chart of a method for tracking and detecting an upper edge of a pantograph pan by a single pixel, and the overall detection method includes the following steps:
firstly, correcting trapezoidal distortion by applying a geometric correction technology to an acquired pantograph image;
secondly, obtaining the area of the pantograph slide plate in the image by adopting an integral chart calculation mode;
thirdly, Canny edge detection is carried out on the area where the sliding plate is located, and an edge image of the pantograph sliding plate is obtained;
fourthly, positioning the lower edges of the contact net and the pantograph slide plate by adopting a Hough detection technology;
and fifthly, traversing upwards from the tail end of the lower edge of the pantograph to find the tail end of the upper edge, and performing single-pixel tracking towards the direction of a contact net to realize complete detection of the upper edge of the pantograph.
The invention is further illustrated by the following figures in conjunction with specific examples.
Fig. 3(a) and 3(b) are images of the left and right sides of the pantograph pan taken by the cameras 1 and 2 at the same height and slightly higher than the pantograph in fig. 1, respectively, in oblique top view. Fig. 4(a) and 4(b) are left and right checkerboard images captured by the cameras 1 and 2 in fig. 1. Fig. 4(c), 4(d), 4(e) and 4(f) are the results obtained after the first step of calibration. Fig. 5(a) and 5(b) are diagrams showing results of positioning the left and right slide plate regions, respectively. Fig. 6(a) and 6(b) show left and right slider edge extraction images. Fig. 7(a) and 7(b) are respectively a left slide plate lower edge and catenary positioning image and a right slide plate lower edge and catenary positioning image. Fig. 8(a), 8(b) are left upper edge localization image and right upper edge localization image. FIG. 9 is a schematic diagram of the manner of integral map calculation. FIG. 10 is a schematic view of a right sled tracking orientation.
Some steps of the present invention are explained and described in more detail below.
In the first step, the invention adopts a perspective transformation method to correct the image. Manually selecting 4 x 4 from the checkerboard image, wherein the size of the checkerboard image is 4 x 4, the area can be freely determined according to the checkerboard area contained in the actual image, and theoretically, the larger the area is, the better the area is; because of trapezoidal distortion, the relation between the pixel value and the actual distance is obtained by taking the vertical direction as reference, and the four vertexes are mapped into the four vertexes of the square area, and the general transformation formula is as follows:
in this embodiment, a picture acquired by any one of the cameras 1 or 2 is selected as an original picture for explanation, and the same processing is performed on a picture acquired by the other camera. (x, y) is the single coordinate of the original picture, and the homogeneous coordinate is obtained after the transformation and the correspondence
[hij](i, j ═ 1,2,3) is a perspective transformation matrix, and the true coordinates (x ', y') are
The conversion formula is arranged to obtain:
it can be seen that in the perspective transformation equation, each matching point pair can contribute two solution equations, four calculated vertexes are target vertexes, and a mapping transformation matrix is obtained and applied to the pantograph image to obtain the pantograph image without geometric distortion
In the second step, the area of the pantograph slide plate in the image obtained by adopting an integral diagram calculation mode is adopted. And 2448 and 300 strip-shaped sliding windows are defined, traversal is started from the top of the image, the step length is 50, the sum of pixels in the sliding windows is calculated by adopting an integral graph mode, and the area where the pixels and the minimum sliding window are located is the area where the target sliding plate is located. FIG. 9 is a diagram illustrating the manner of integral map calculation used in the second step of the present invention. The integral graph calculation formula of the position (i, j) of a certain pixel point in the image is as follows:
I(i,j)＝I(i,j-1)+I(i-1,j)-I(i-1,j-1)+image(i,j)
wherein I (I, j) represents the integral of the pixel point (I, j), and image (I, j) represents the gray value of the pixel point (I, j).
In the third step, the invention adopts a Canny algorithm to detect the edge of the pantograph slide plate. The method comprises the following specific steps:
first, the image is convolved in the x and y directions with the following convolution operators to compute the image gradient values:
two partial derivative matrices of the image in the x and y directions can be obtained by using a first order finite difference approximation to calculate the gradient. Due to the gradient vector, the magnitude and angle of the gradient are calculated as follows:
P[i,j]＝(f[i,j+1]-f[i,j]+f[i+1,j+1]-f[i+1,j])/2
Q[i,j]＝(f[i,j]-f[i+1,j]+f[i,j+1]-f[i+1,j+1])/2
θ[i,j]＝arctan(Q[i,j]/P[i,j])
wherein M [ i, j ] is the amplitude of the gradient, θ [ i, j ] is the direction of the gradient, and f (i, j) is the gray value of the pixel point (i, j).
Then, after the gradient amplitude and the angle of the point C are calculated, in order to determine whether the point C is an edge of the image, the pixel value of the point C is compared with the pixel values in the 8 neighborhoods thereof, and if the gray value of the point C is not the maximum, it is indicated that the point C is not an edge point of the image, and the gray value of the point C can be assigned as 0. After this step is completed, a binary image containing image edge information is obtained, the non-edge points have been set to 0, and the gray value of the points that may be edges can be assigned to 255.
Then, the maximum pixel value Max and the minimum pixel value Min of the image are found, and the high threshold value is defined as TH0.7Max, low threshold TLIf the gradient value of the pixel point is higher than the high threshold, the point is determined to be an edge point, if the gradient value of the pixel point is lower than the low threshold, the point is determined to be a non-edge point, and if the gradient value of the pixel point is between the high threshold and the low threshold. And further inspecting the gradient value of the pixel point in the neighborhood of the pixel point 8, if the gradient value of the peripheral pixel point is higher than the high threshold value, determining the pixel point as an edge point, and if the gradient values of the peripheral pixel points are all smaller than the high threshold value, determining the pixel point as a non-edge point.
This step acquires an image of the edge of the pantograph slide plate.
In the fourth step, the lower edge of the pantograph slide plate and a contact net are positioned by Hough transformation.
First, straight lines in an image are detected using the Hough transform line detection technique. And if the slope of the straight line is smaller than a first set value (the value of the slope is related to the shooting angle and the shooting position of the camera, and 0.12 is the best value under a specific shooting condition), and the coordinates of the bottom of the target sliding plate area determined in the approximation step II are within a set pixel range, the straight line is considered to be the lower edge of the pantograph sliding plate, and if the slope of the straight line is between a second set value and a third set value, the straight line is considered to be the pantograph of the pantograph. In this embodiment, specific values are as follows:
if the slope k of the straight line is less than 0.12 and the coordinates of the bottom of the target sliding plate area determined in the approximation step II are within +/-50 pixels, the straight line is considered as the lower edge of the pantograph sliding plate, and the pixel where the straight line is located is given as yellow; if k is more than 0.2 and less than 0.4, the straight line is considered as the pantograph contact network, and the pixel where the straight line is located is given pink.
The setting is related to the height and angle set by the camera, and in this embodiment, the cameras 1 and 2 are at the same height and slightly higher than the pantograph, and the oblique photographing mode is adopted in the top view, so the values are taken as the above-mentioned condition, and are not limited to the angle and pixel range adopted in this embodiment.
In the fifth step, the invention adopts a single-pixel tracking detection method for the upper edge to position the upper edge of the pantograph slide plate.
The end of the lower edge of the pantograph slide positioned above is first found. Because the contact abrasion of the pantograph slide plate is usually concentrated in the center of the pantograph slide plate, the tail end of the pantograph slide plate is usually the best state that the pantograph slide plate is not abraded, taking the right half-slide plate of the pantograph as an example, the first pixel point with the pixel value of 255 is searched upwards from the tail end of the lower edge of the right lower slide plate, namely the tail end position (i, j) of the upper edge of the pantograph slide plate.
And in view of the fact that only pixel points with pixel values of only 0 and 255 are left in the pantograph image after the edge detection, searching from the upper edge end of the pantograph to the pantograph overhead line system. FIG. 10 is a schematic view of the tracking orientation of the right slide used in the fifth step of the present invention. Taking the right pantograph pan half as an example, only the positions (i-1, j-1), (i-1, j) and (i-1, j +1) are searched, because of the continuity of the edge of the pantograph pan, the three positions must have a pixel point with a pixel value of 255, downward search is continued until the pixel value is 0, downward search is stopped, and the last position with the pixel value of 255 is the upper edge of the pantograph pan.
And repeating the steps until the position of the overhead line system is reached.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
The above-mentioned embodiments, objects, technical solutions and advantages of the present invention are further described in detail, it should be understood that the above-mentioned embodiments are only illustrative of the present invention and are not intended to limit the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (5)
1. A pantograph slide plate upper edge single-pixel tracking detection method is characterized by comprising the following steps:
step 1: correcting trapezoidal distortion by applying a geometric correction technology to the acquired pantograph image; the detailed steps of the geometric correction in the step 1 are as follows:
step 1.1: when the pantograph slide plate does not pass through, vertically placing a calibration checkerboard at the position of the photoelectric sensor, triggering an industrial camera, and acquiring a checkerboard image;
step 1.2: manually selecting four vertexes of a rectangular checkerboard area in the checkerboard image, obtaining a relation between a pixel value and an actual distance by taking a vertical direction as a reference, mapping the four vertexes of the checkerboard area into the four vertexes of the rectangular area, obtaining the four vertexes as target vertexes through calculation, and obtaining a mapping transformation matrix through calculation;
step 1.3: applying the transformation matrix to the pantograph image to obtain a pantograph image without geometric distortion;
step 2: obtaining the area of the pantograph slide plate in the image by adopting an integral chart calculation mode;
and step 3: carrying out Canny edge detection on the area where the sliding plate is located to obtain an edge image of the pantograph sliding plate;
and 4, step 4: positioning the lower edges of the contact net and the pantograph slide plate by using a Hough detection technology;
and 5: and traversing upwards from the tail end of the lower edge of the pantograph to find the tail end of the upper edge, and performing single-pixel tracking towards the direction of a contact net to realize complete detection of the upper edge of the pantograph.
2. The method for detecting the upper edge of the pantograph pan by single-pixel tracking according to claim 1, wherein the integral graph calculation mode in the step 2 comprises the following detailed steps:
step 2.1: defining a strip sliding window, and traversing from the top of the image;
step 2.2: and calculating the sum of pixels in the sliding window by adopting an integral graph mode, wherein the area where the pixel and the minimum sliding window are located is the area where the target sliding plate is located.
3. The method according to claim 1, wherein the Canny edge detection algorithm in step 3 comprises the following detailed steps:
step 3.1: the image is convolved in the x and y directions with the following convolution operators to compute the image gradient values:
two partial derivative matrixes of the image in the x direction and the y direction are obtained by a method of calculating the gradient by first-order finite difference approximation, and the amplitude and the angle of the gradient are calculated as follows due to the gradient vector:
P[i,j]＝(f[i,j+1]-f[i,j]+f[i+1,j+1]-f[i+1,j])/2
Q[i,j]＝(f[i,j]-f[i+1,j]+f[i,j+1]-f[i+1,j+1])/2
θ[i,j]＝arctan(Q[i,j]/P[i,j])
wherein M [ i, j ] is the amplitude of the gradient, theta [ i, j ] is the direction of the gradient, and f (i, j) is the gray value of the pixel point (i, j);
step 3.2: after the gradient amplitude and the angle of the whole image are calculated in the step 3.1, for any pixel point C in the image, in order to determine whether the point C is the edge of the image, comparing the pixel value of the point C with the pixel values in the neighborhood of the point C, if the gray value of the point C is not the maximum value, indicating that the point C is not the edge point of the image, and assigning the gray value of the point C as 0;
after the step is finished, a binary image containing image edge information is obtained, non-edge points are set to be 0, and gray values of the points which are possibly edges can be assigned to be 255;
step 3.3: finding out the maximum pixel value Max and the minimum pixel value Min of the image, defining a high threshold value and a low threshold value, if the gradient value of the pixel point is higher than the high threshold value, judging the point as an edge point, and if the gradient value of the pixel point is lower than the low threshold value, judging the point as a non-edge point;
if the gradient value of the pixel point is between the high threshold and the low threshold, the gradient value of the pixel point in the neighborhood of the pixel point is further inspected, if the peripheral pixel points with the gradient values higher than the high threshold exist, the pixel point is still judged as an edge point, and if the gradient values of the peripheral pixel points are smaller than the high threshold, the pixel point is judged as a non-edge point.
4. The method for detecting the upper edge of the pantograph pan by single-pixel tracking according to claim 2, wherein the Hough transform positions the lower edge and the overhead line system in the step 4, and the method comprises the following specific steps:
step 4.1: detecting a straight line in the image by using a Hough transformation line detection technology;
step 4.2: and if the slope of the straight line is smaller than the first set value and the bottom coordinate determined by the target sliding plate area method determined in the step two is set in the pixel range, the straight line is considered to be the lower edge of the pantograph sliding plate, and if the slope of the straight line is between the second set value and the third set value, the straight line is considered to be the pantograph contact net.
5. The method for detecting the upper edge of the pantograph pan by single-pixel tracking according to claim 1, wherein the step 5 comprises the following steps:
step 5.1: firstly, finding out the tail end for positioning the lower edge of the pantograph slide plate, wherein the contact abrasion of the pantograph slide plate is usually concentrated in the center of the pantograph slide plate, and the tail end is the best state that the slide plate is not abraded;
step 5.2: in view of the fact that only pixel points with pixel values of only 0 and 255 are left in the pantograph image after edge detection, searching from the end of the upper edge of the pantograph to the direction of a pantograph catenary; searching positions (i-1, j-1), (i-1, j) and (i-1, j +1) by a right pantograph half skateboard, searching positions (i +1, j-1), (i +1, j) and (i +1, j +1) by a left pantograph half skateboard, wherein due to the continuity of the edge of the pantograph slide board, pixel points with the pixel value of 255 are necessarily arranged in the three positions, continuously searching downwards until the pixel value is 0, stopping searching downwards, and determining the position with the last pixel value of 255 as the upper edge of the pantograph slide board;
step 5.3: and repeating the steps until the position of the overhead line system is reached.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810699726.7A CN108917621B (en) | 2018-06-29 | 2018-06-29 | Pantograph slide plate upper edge single-pixel tracking detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810699726.7A CN108917621B (en) | 2018-06-29 | 2018-06-29 | Pantograph slide plate upper edge single-pixel tracking detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108917621A CN108917621A (en) | 2018-11-30 |
| CN108917621B true CN108917621B (en) | 2020-08-07 |
Family
ID=64423457
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810699726.7A Active CN108917621B (en) | 2018-06-29 | 2018-06-29 | Pantograph slide plate upper edge single-pixel tracking detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108917621B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109961429A (en) * | 2019-03-22 | 2019-07-02 | 武汉工程大学 | A kind of pantograph detection localization method and system based on monocular infrared image |
| CN110910443B (en) * | 2019-12-04 | 2023-03-21 | 成都唐源电气股份有限公司 | Contact net geometric parameter real-time measuring method and device based on single monitoring camera |
| CN112781493B (en) * | 2020-12-25 | 2022-09-06 | 江苏集萃智能光电系统研究所有限公司 | Online detection equipment for comprehensive geometric parameters of pantograph of train |
| CN114973089A (en) * | 2022-05-30 | 2022-08-30 | 福州大学 | Contact net and pantograph contact point detection method based on image vision algorithm |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107588733A (en) * | 2017-08-21 | 2018-01-16 | 南京理工大学 | A kind of abrasion of pantograph pan on-line measuring device and method based on image |
| CN107742288A (en) * | 2017-10-11 | 2018-02-27 | 成都国铁电气设备有限公司 | A kind of pantograph beats bow defect inspection method and system |
-
2018
- 2018-06-29 CN CN201810699726.7A patent/CN108917621B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN108917621A (en) | 2018-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108917621B (en) | Pantograph slide plate upper edge single-pixel tracking detection method | |
| CN107014294B (en) | Contact net geometric parameter detection method and system based on infrared image | |
| CN112418103B (en) | Bridge crane hoisting safety anti-collision system and method based on dynamic binocular vision | |
| WO2015024407A1 (en) | Power robot based binocular vision navigation system and method based on | |
| CN107703513B (en) | Non-contact net relative position detection method based on image processing | |
| US20190295241A1 (en) | Inspection device, control method and control apparatus for the same | |
| CN105203552A (en) | 360-degree tread image detecting system and method | |
| CN109300125B (en) | Method for extracting pantograph slide plate image of urban rail train in complex environment | |
| CN108725511B (en) | Real-time position correction method for rail corrugation measuring point | |
| KR102010823B1 (en) | Method and apparatus for measuring speed of vehicle by using fixed single camera | |
| CN106996748A (en) | A kind of wheel footpath measuring method based on binocular vision | |
| JP2017083245A (en) | Clearance limit determination device | |
| Espino et al. | Rail extraction technique using gradient information and a priori shape model | |
| CN108564627B (en) | Linear array image sleeper positioning and counting method based on multi-region gray projection | |
| KR20230073221A (en) | Tread depth estimation system and method thereof | |
| CN111127381B (en) | Non-parallel detection method for pantograph slide plate | |
| CN113567451A (en) | Cable defect detection and diameter measurement method | |
| Kiddee et al. | Visual recognition of the initial and end points of lap joint for welding robots | |
| Wang et al. | Geometry constraints-based visual rail track extraction | |
| Di Stefano et al. | Automatic 2D-3D vision based assessment of the attitude of a train pantograph | |
| CN113409303A (en) | Target detection and tracking method for high-speed railway contact network | |
| CN110322508B (en) | Auxiliary positioning method based on computer vision | |
| JP5952759B2 (en) | Overhead wire position measuring apparatus and method | |
| US20190139177A1 (en) | Device for detecting road surface state | |
| CN117369460A (en) | Intelligent inspection method and system for loosening faults of vehicle bolts |
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 |