CN108917621A - A kind of pantograph pan top edge single pixel tracking detection method - Google Patents
A kind of pantograph pan top edge single pixel tracking detection method Download PDFInfo
- Publication number
- CN108917621A CN108917621A CN201810699726.7A CN201810699726A CN108917621A CN 108917621 A CN108917621 A CN 108917621A CN 201810699726 A CN201810699726 A CN 201810699726A CN 108917621 A CN108917621 A CN 108917621A
- Authority
- CN
- China
- Prior art keywords
- pantograph
- pixel
- image
- edge
- value
- 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
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 kind of pantograph pan top edge single pixel tracking detection methods, geometric correction is carried out to pantograph image by perspective transform first, then pantograph pan region in image is found using integrogram calculation, Canny edge detection is carried out to pantograph image again, on this basis using Hough transform positioning contact net and by slide plate lower edge, pantograph pan top edge is finally positioned using single pixel tracking detection method.The technology can efficiently obtain pantograph pan effective information, improve the accuracy of pantograph pan marginal information, have stronger practicability and wide application prospect.
Description
Technical field
The present invention relates to the online automatic detection technical fields of rail vehicle, refer in particular to the machines such as railway, urban rail and subway
Vehicle car pantograph abrasion value detection field, specifically a kind of pantograph pan top edge single pixel tracking detection method.
Background technique
Can pantograph be the important device that locomotive removes electricity from power grid, work normally the row that will have a direct impact on locomotive
Sail safety.The abrasion situation of pantograph pan is to influence an important factor for pantograph operates normally.Therefore, it accurately detects by electricity
The wearing valve for bending slide plate has important reference significance to instructing locomotive depot personnel to climb to the top of a mountain to overhaul.
Electric locomotive in the process of running, needs the contact by pantograph pan with contact net, electricity is taken from power grid.This
The mode of kind contact power supply inevitably results in the electrical abrasion and sliding abrasion of pantograph pan.If letting alone abrasion without taking
Action, it will cause seine even card nets, and What is more becomes the immediate cause of serious railway traffic accident.
Conventional method is that abrasion of pantograph pan value is detected using manual measurement, it is desirable that parking detection, it is time-consuming and laborious and
There are manual measurement errors.With the development of image processing techniques, the ruler of object is carried out by the high-definition picture acquired
The detection techniques such as very little, shape grow up and are widely applied.Image detecting technique has:It is contactless, detection is accurate, detection effect
The advantages that rate is high.In image detection, can pantograph pan top edge accurately be detected, and become influence pantograph pan mill
The important factor in order of consumption value detection accuracy.Therefore, a kind of high-precision, the pantograph edge detection method of high reliability are studied
It is significant to the detection level for improving abrasion of pantograph pan of locomotive.
Summary of the invention
The technical problem to be solved by the present invention is to position for existing abrasion of pantograph pan value context of detection top edge
The deficiency of aspect proposes a kind of pantograph pan top edge single pixel tracking detection method using image processing techniques, can
Efficiently and accurately position pantograph pan top edge.
The present invention uses following technical scheme to solve above-mentioned technical problem:
A kind of pantograph pan top edge single pixel tracking detection method, includes the following steps:
Step 1:Geometric correction technology is used to pantograph image, corrects trapezoidal distortion;
Step 2:By the way of integrogram calculating, pantograph pan region in obtained image;
Step 3:Canny edge detection is carried out to slide plate region, obtains pantograph pan edge graph;
Step 4:Contact net and pantograph pan lower edge are positioned using Hough detection technique;
Step 5:It is traversed up from pantograph lower edge end and seeks obtaining top edge end, carry out single pixel to contact net direction
Tracking realizes that pantograph top edge is completely detected.
As a kind of further prioritization scheme of pantograph pan top edge single pixel tracking detection method of the invention, step
It is as follows to the detailed step of pantograph image progress geometric correction in 1:
Step 1.1:When no pantograph pan passes through, calibration chessboard is placed vertically in photoelectric sensor position
Lattice trigger industrial camera, obtain chessboard table images;
Step 1.2:Four vertex for choosing 4*4 gridiron pattern region manually in chessboard table images, since there are trapezoidal mistakes
Very, therefore with vertical direction it is reference, obtains the relationship between pixel value and actual range, four vertex mappings is square
Four vertex in region, counting counted four vertex is representative points.Projective transformation matrix is obtained by calculating.
Step 1.3:Transformation matrix is applied to pantograph image, obtains the pantograph image without geometric distortion.
As a kind of further prioritization scheme of pantograph pan top edge single pixel tracking detection method of the invention, step
It is as follows to the detailed step of pantograph image progress slide plate zone location in 2:
Step 2.1:The strip sliding window for defining 2448*300, begins stepping through, step-length 50 from image the top;
Step 2.2:Using integrogram mode calculate in sliding window pixel and, pixel and the smallest sliding window region are
For target slide plate region.
As a kind of further prioritization scheme of pantograph pan top edge single pixel tracking detection method of the invention, step
The detailed step of Canny edge detection is as follows in 3:
Step 3.1:Carry out convolution in the x and y direction to image with convolution operator below to calculate image gradient value:
It is divided to the method for approximate calculation gradient to can get the two partial derivative squares of image in the x and y direction with first difference
Battle array.It is as follows to the amplitude and angle computation method of gradient due to 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 gradient, and θ [i, j] is the direction of gradient.
Step 3.2:After the gradient magnitude and the angle that calculate point C, for determine point C whether be image edge, by C point picture
Plain value is compared with the pixel value in its 8 neighborhood, if the gray value of C is not wherein maximum, illustrates that C is not just image border
The gray value of C can be assigned to 0 by point.The bianry image comprising image edge information is obtained after completing this step, non-edge point is
It is set to 0, its gray value can may be assigned to 255 for the point at edge;
Step 3.3:Image maximum pixel value Max and minimum pixel value Min are found out, definition high threshold is TH=0.7Max,
Low threshold is TL=0.3Min determines that the point is marginal point, if pixel gradient value if pixel gradient value is higher than high threshold
Lower than Low threshold, then determine that the point must be non-edge point, if pixel gradient value is between high-low threshold value.Then further investigate
The gradient value size of the 8 neighborhood territory pixel point of pixel, if periphery has gradient value to be higher than the pixel of high threshold, the pixel is still
It is determined as marginal point, if peripheral image vegetarian refreshments gradient value is respectively less than high threshold, which is determined as non-edge point.
As a kind of further prioritization scheme of pantograph pan top edge single pixel tracking detection method of the invention, step
The detailed step of Hough detection technique is as follows in 4:
Step 4.1:Use the straight line in Hough transform line detection technique detection image.
Step 4.2:If straight slope k < 0.12 (value of slope is related with camera shooting angle and camera site, and 0.12
It is the best value in the case of specific shooting), and ± 50 picture of target slide plate sections bottom coordinate determined in approximating step two
In plain range, then it is assumed that the straight line is pantograph pan lower edge, if 0.2 < k < 0.4, then it is assumed that the straight line is that pantograph connects
It touches net.
As a kind of further prioritization scheme of pantograph pan top edge single pixel tracking detection method of the invention, step
The detailed step of top edge track algorithm is as follows in 5:
Step 5.1:The end of positioning pantograph pan lower edge above is found out first.Since pantograph pan contact is ground
Damage often focuses on the center of pantograph pan, and end is often the optimum state that slide plate does not generate abrasion, with pantograph right half
For slide plate, since the square slide lower edge end of bottom right, the pixel that pixel value is 255 at first is found in search upwards, i.e.,
For the terminal position (i, j) of pantograph pan top edge.
Step 5.2:In view of in the pantograph image after edge detection only remaining pixel value there was only 0, the picture of 255 two kind of situation
Vegetarian refreshments is searched for from pantograph top edge end toward pantograph contact net direction.Half slide plate of the pantograph right side searches for (i-1, j-1), (i-
1, j) and the position (i-1, j+1), left half slide plate search for the position (i+1, j-1), (i+1, j) and (i+1, j+1), since pantograph is sliding
The continuity of edges of boards edge, must have pixel value in these three positions is 255 pixel, continues search downwards until pixel value is 0,
Stop search downwards, the position that the last one pixel value is 255 is then pantograph pan top edge.
Step 5.3:It repeats the above steps, until reaching contact net position.
The invention adopts the above technical scheme compared with prior art, has the following technical effects:
A kind of pantograph pan top edge single pixel tracking detection method proposed by the present invention, first to pantograph image into
Row geometric correction, then positions target slide plate region using integrogram calculation, then with Canny edge detection to by
Pantograph slider image carries out edge extracting, on this basis Hough line detection technique positioning slide plate lower edge and contact net, finally
Pantograph pan top edge is positioned using top edge track algorithm, can efficiently obtain pantograph pan effective information, improve by
The accuracy of pantograph slider marginal information has stronger practicability and wide application prospect.
Detailed description of the invention
Fig. 1 is image capturing system schematic layout pattern;
Fig. 2 is that a kind of process of pantograph pan top edge single pixel tracking detection method of certain embodiments of the invention is shown
It is intended to;
Fig. 3 (a), Fig. 3 (b) are the first acquired original image schematic diagram and the second acquired original image schematic diagram respectively;
Fig. 4 (a), Fig. 4 (b), Fig. 4 (c), Fig. 4 (d), Fig. 4 (e), Fig. 4 (f) are the first chessboard table images, the second chess respectively
The gridiron pattern several picture after chessboard table images, the second correction, the first acquired original figure after disk table images, the first geometric correction
As the image after the image and the second acquired original image rectification after correction;
Fig. 5 (a), 5 (b) are the first correction tail skid zone location image and the second correction tail skid regional map respectively
Picture;
Fig. 6 (a), 6 (b) are the first correction tail skid edge extracting image and the second correction tail skid edge extracting figure respectively
Picture;
Fig. 7 (a), 7 (b) are the first correction tail skid lower edge and contact net positioning image and the second correction tail skid respectively
Lower edge and contact net position image;
Fig. 8 (a), 8 (b) are that the first correction tail skid uses top edge positioning image and the second school after single pixel tracking respectively
Positive tail skid positions image using top edge after single pixel tracking;
Fig. 9 is integrogram calculation schematic diagram;
Figure 10 is right slide plate tracking orientation schematic diagram.
Specific embodiment
Technical solution of the present invention is described in further detail with reference to the accompanying drawing:
Fig. 1 show image capturing system schematic layout pattern employed in the present embodiment, is motorcycle pantograph abrasion inspection
The citing of the exemplary laying mode of examining system, wherein image capturing system is used to acquire the realtime graphic of motorcycle pantograph, packet
Include Image Acquisition camera, infrared triggering switch, the light source etc. being laid in above track.
Fig. 2 show a kind of flow diagram of pantograph pan top edge single pixel tracking detection method, generally whole
A detection method includes following processes:
The first step uses geometric correction technology to the pantograph image of acquisition, corrects trapezoidal distortion;
Second step, by the way of integrogram calculating, pantograph pan region in obtained image;
Third step carries out Canny edge detection to slide plate region, obtains pantograph pan edge graph;
4th step positions contact net and pantograph pan lower edge using Hough detection technique;
5th step traverses up from pantograph lower edge end and seeks obtaining top edge end, carries out single picture to contact net direction
Element tracking realizes that pantograph top edge is completely detected.
Below with reference to diagram, the invention will be further described with specific example.
Fig. 3 (a), Fig. 3 (b) are that camera 1,2 takes vertical view tiltedly to clap in sustained height and slightly above pantograph in Fig. 1 respectively
Mode, on the left of the pantograph pan of shooting, image right.Fig. 4 (a), Fig. 4 (b) are the shooting of camera 1,2 left side, right side in Fig. 1
Chessboard table images.Fig. 4 (c), Fig. 4 (d), Fig. 4 (e), Fig. 4 (f) are acquired results after step 1 correction.Fig. 5 (a), 5 (b) are respectively
It is left and right sideskid zone location result figure.Fig. 6 (a), 6 (b) obtain left and right slide plate edge and extract image.Fig. 7 (a), 7 (b) are respectively
It is left hand ski lower edge and contact net positioning image and right hand ski lower edge and contact net positioning image.Fig. 8 (a), 8 (b)
It is left side top edge positioning image and right side top edge positioning image.Fig. 9 is integrogram calculation schematic diagram.Figure 10 is right cunning
Plate tracks orientation schematic diagram.
The some steps of the present invention are explained in more detail and are described below.
In the first step, the present invention carries out image rectification using the method for perspective transform.In chessboard table images manually
4*4 is chosen, the present embodiment is used having a size of 4*4, which can freely determine according to the gridiron pattern region for including in real image,
Theoretically it is the bigger the better;Four vertex in gridiron pattern region are reference with vertical direction, obtain since there are trapezoidal distortions
The relationship between pixel value and actual range is obtained, four vertex are mapped as to four vertex of square area, general transformation
Formula is:
1, the 2 collected picture of institute of camera is required to carry out image flame detection, in the present embodiment, chooses camera 1 or 2
One picture collected of meaning is illustrated as original image, another collected picture of camera will also be carried out identical
Processing.(x, y) is the single coordinate of original image, and correspondence obtains homogeneous coordinates after transformation[hij] (i, j=1,2,3) be
Perspective transformation matrix, real coordinate (x ', y ') areIt is available to arrange transformation for mula:
As can be seen that each matching double points can contribute two solution equations in perspective transform equation, counted four are counted
Vertex is representative points, and obtains projective transformation matrix, is applied to pantograph image, obtains the pantograph without geometric distortion
Image
In second step, pantograph pan region in the image that the present invention uses integrogram calculation to obtain.Definition
The strip sliding window of 2448*300, begins stepping through from image the top, and step-length 50 is calculated in sliding window using integrogram mode
Pixel and, pixel and the smallest sliding window region are target slide plate region.Fig. 9 is made by second step in the present invention
The schematic diagram of integrogram calculation.The integrogram calculation formula of the position (i, j) of a certain pixel is as follows in image:
I (i, j)=I (i, j-1)+I (i-1, j)-I (i-1, j-1)+image (i, j)
Wherein I (i, j) indicates that the integral of pixel (i, j), image (i, j) indicate the gray value of vegetarian refreshments (i, j).
In third step, the present invention takes Canny algorithm to detect pantograph pan edge.Specific step is as follows for it:
Firstly, carrying out convolution in the x and y direction to image with convolution operator below to calculate image gradient value:
It is divided to the method for approximate calculation gradient to can get the two partial derivative squares of image in the x and y direction with first difference
Battle array.It is as follows to the amplitude and angle computation method of gradient due to 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 gradient, and θ [i, j] is the direction of gradient, and f (i, j) is the gray scale of pixel (i, j)
Value.
Then, after the gradient magnitude and the angle that calculate point C, for determine point C whether be image edge, by C point pixel
Value is compared with the pixel value in its 8 neighborhood, if the gray value of C be not it is wherein maximum, illustrating C just is not image border point,
The gray value of C can be assigned to 0.Complete to obtain the bianry image comprising image edge information after this step, non-edge point by
It is set to 0, its gray value can may be assigned to 255 for the point at edge.
Then, image maximum pixel value Max and minimum pixel value Min are found out, definition high threshold is TH=0.7Max, low threshold
Value is TL=0.3Min determines that the point is marginal point, if pixel gradient value is lower than if pixel gradient value is higher than high threshold
Low threshold then determines that the point must be non-edge point, if pixel gradient value is between high-low threshold value.Then further investigate the picture
The gradient value size of 8 neighborhood territory pixel point of vegetarian refreshments, if the pixel that periphery has gradient value to be higher than high threshold, the pixel still determine
For marginal point, if peripheral image vegetarian refreshments gradient value is respectively less than high threshold, which is determined as non-edge point.
The step obtains pantograph pan edge image.
In 4th step, the present invention is using Hough transform positioning pantograph pan lower edge and contact net.
First using the straight line in Hough transform line detection technique detection image.If straight slope is less than the first setting value
(value of slope is related with camera shooting angle and camera site, and 0.12 is the best value in the case of specific shooting), and
In the target slide plate sections bottom setting coordinate pixel coverage determined in approximating step two, then it is assumed that the straight line is pantograph pan
Lower edge, if straight slope is between second and third setting value, then it is assumed that the straight line is pantograph contact net.The present embodiment
In, specific value is as follows:
If straight slope k < 0.12, and ± 50 pixel of target slide plate sections bottom coordinate determined in approximating step two
In range, then it is assumed that the straight line is pantograph pan lower edge, and pixel where the straight line is assigned to yellow;If 0.2 < k < 0.4,
Then think that the straight line is pantograph contact net, pixel where the straight line is assigned to pink colour.
The setting and and camera setting height it is related with angle, in the present embodiment, camera 1,2 is in sustained height and slightly higher
In pantograph, take the mode overlooked and tiltedly clapped, therefore value is above situation, however it is not limited to angle employed in the present embodiment with
Pixel coverage.
In 5th step, the present invention positions pantograph pan top edge using top edge single pixel tracking detection method.
The end of positioning pantograph pan lower edge above is found out first.Since the abrasion of pantograph pan contact often collects
In at the center of pantograph pan, end is often the optimum state that slide plate does not generate abrasion, by taking right half slide plate of pantograph as an example,
Since the square slide lower edge end of bottom right, the pixel that pixel value is 255 at first is found in search upwards, and as pantograph is sliding
The terminal position (i, j) of plate top edge.
In view of in the pantograph image after edge detection only remaining pixel value there was only 0, the pixel of 255 two kind of situation, from
Pantograph top edge end is searched for toward pantograph contact net direction.Figure 10 is right slide plate tracking used in the 5th step in the present invention
Orientation schematic diagram.By taking right half slide plate of pantograph as an example, search only for the position (i-1, j-1), (i-1, j) and (i-1, j+1), due to by
The continuity at pantograph slider edge, must have pixel value in these three positions is 255 pixel, continues search downwards until pixel
Value is 0, stops search downwards, the position that the last one pixel value is 255 is then pantograph pan top edge.
It repeats the above steps, until reaching contact net position.
Those skilled in the art can understand that unless otherwise defined, all terms used herein (including skill
Art term and scientific term) there is meaning identical with the general understanding of those of ordinary skill in fields of the present invention.Also
It should be understood that those terms such as defined in the general dictionary should be understood that have in the context of the prior art
The consistent meaning of meaning will not be explained in an idealized or overly formal meaning and unless defined as here.
Above-described specific embodiment has carried out further the purpose of the present invention, technical scheme and beneficial effects
It is described in detail, it should be understood that being not limited to this hair the foregoing is merely a specific embodiment of the invention
Bright, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should be included in the present invention
Protection scope within.
Claims (6)
1. a kind of pantograph pan top edge single pixel tracking detection method, which is characterized in that step includes:
Step 1:Geometric correction technology is used to the pantograph image of acquisition, corrects trapezoidal distortion;
Step 2:By the way of integrogram calculating, pantograph pan region in obtained image;
Step 3:Canny edge detection is carried out to slide plate region, obtains pantograph pan edge graph;
Step 4:Contact net and pantograph pan lower edge are positioned using Hough detection technique;
Step 5:Traversed up from pantograph lower edge end and seek obtaining top edge end, to contact net direction carry out single pixel with
Track realizes that pantograph top edge is completely detected.
2. a kind of pantograph pan top edge single pixel tracking detection method according to claim 1, it is characterised in that described
The detailed step of geometric correction is as follows in step 1:
Step 1.1:When no pantograph pan passes through, calibration gridiron pattern, touching are placed vertically in photoelectric sensor position
Industrial camera is sent out, chessboard table images are obtained;
Step 1.2:It chooses four vertex in rectangle gridiron pattern region manually in chessboard table images, is reference with vertical direction,
The relationship between pixel value and actual range is obtained, four vertex of gridiron pattern are mapped as to four vertex of rectangular area, is calculated
Four obtained vertex are representative points, then obtain projective transformation matrix by calculating;
Step 1.3:Transformation matrix is applied to pantograph image, obtains the pantograph image without geometric distortion.
3. a kind of pantograph pan top edge single pixel tracking detection method according to claim 1, it is characterised in that described
Integrogram calculation in step 2, detailed step are as follows:
Step 2.1:Strip sliding window is defined, is begun stepping through from image the top;
Step 2.2:Using integrogram mode calculate in sliding window pixel and, pixel and the smallest sliding window region are mesh
Mark slide plate region.
4. a kind of pantograph pan top edge single pixel tracking detection method according to claim 1, it is characterised in that described
Canny edge detection algorithm described in step 3, detailed step are as follows:
Step 3.1:Carry out convolution in the x and y direction to image with convolution operator below to calculate image gradient value:
It is divided to the method for approximate calculation gradient to obtain two partial derivative matrixs of image in the x and y direction with first difference, due to
Gradient vector, it is as follows to the amplitude and angle computation method of gradient:
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 gradient, and θ [i, j] is the direction of gradient, and f (i, j) is the gray value of pixel (i, j);
Step 3.2:After the gradient magnitude and the angle that calculate entire image in step 3.1, for pixel any one in image
Point C, for determine point C whether be image edge, C point pixel value is compared with the pixel value in its neighborhood, if the gray scale of C
Value be not it is wherein maximum, illustrating C not is image border point, and the gray value of C is assigned to 0;
The bianry image comprising image edge information is obtained after completing this step, non-edge point has been set to 0, may be edge
Its gray value can be assigned to 255 by point;
Step 3.3:Image maximum pixel value Max and minimum pixel value Min are found out, high threshold, Low threshold are defined, if pixel is terraced
Angle value is higher than high threshold, then determines that the point is marginal point, if pixel gradient value is lower than Low threshold, determines that the point must be non-side
Edge point;
If pixel gradient value, between high-low threshold value, the gradient value for further investigating the neighborhood of pixel points pixel is big
Small, if periphery has gradient value to be higher than the pixel of high threshold, which is still determined as marginal point, if peripheral image vegetarian refreshments gradient
Value is respectively less than high threshold, then the pixel is determined as non-edge point.
5. a kind of pantograph pan top edge single pixel tracking detection method according to claim 3, it is characterised in that described
Hough transform described in step 4 positions lower edge and contact net, and specific step is as follows:
Step 4.1:Use the straight line in Hough transform line detection technique detection image;
Step 4.2:If straight slope is less than the first setting value, and the target slide plate sections bottom determined in approximating step two is sat
In mark setting pixel coverage, then it is assumed that the straight line is pantograph pan lower edge, if straight slope is in second and third setting value
Between, then it is assumed that the straight line is pantograph contact net.
6. a kind of pantograph pan top edge single pixel tracking detection method according to claim 1, it is characterised in that described
Top edge track algorithm described in step 5, specific step is as follows:
Step 5.1:The end for finding out positioning pantograph pan lower edge first, since the abrasion of pantograph pan contact often collects
In at the center of pantograph pan, end is the optimum state that slide plate does not generate abrasion, by taking right half slide plate of pantograph as an example, from the right side
Lower square slide lower edge end starts, and the pixel that pixel value is 255 at first is found in search upwards, as on pantograph pan
The terminal position (i, j) at edge;
Step 5.2:In view of in the pantograph image after edge detection only remaining pixel value there was only 0, the pixel of 255 two kind of situation
Point is searched for from pantograph top edge end toward pantograph contact net direction;The pantograph right side half slide plate search (i-1, j-1), (i-1,
J) with the position (i-1, j+1), left half slide plate searches for the position (i+1, j-1), (i+1, j) and (i+1, j+1), due to pantograph pan
The continuity at edge, must have pixel value in these three positions is 255 pixel, continues search downwards until pixel value is 0, stops
Only search downwards, the position that the last one pixel value is 255 is then pantograph pan top edge;
Step 5.3:It repeats the above steps, until reaching contact net position.
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 true CN108917621A (en) | 2018-11-30 |
CN108917621B 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) |
Cited By (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 |
CN110910443A (en) * | 2019-12-04 | 2020-03-24 | 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 |
CN112781493A (en) * | 2020-12-25 | 2021-05-11 | 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 |
Citations (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
Patent Citations (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 |
Non-Patent Citations (1)
Title |
---|
师鹏燕: "基于仰拍方式的受电弓滑板磨耗检测系统的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (5)
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 |
CN110910443A (en) * | 2019-12-04 | 2020-03-24 | 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 |
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 |
CN112781493A (en) * | 2020-12-25 | 2021-05-11 | 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 |
Also Published As
Publication number | Publication date |
---|---|
CN108917621B (en) | 2020-08-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108917621A (en) | A kind of pantograph pan top edge single pixel tracking detection method | |
CN106291278B (en) | A kind of partial discharge of switchgear automatic testing method based on more vision systems | |
CN104236498B (en) | Method for measuring end size of switch rail | |
CN103714538B (en) | road edge detection method, device and vehicle | |
CN110142785A (en) | A kind of crusing robot visual servo method based on target detection | |
CN107516077A (en) | Traffic sign information extracting method based on laser point cloud and image data fusion | |
US20150248579A1 (en) | Method for identifying and positioning building using outline region restraint of mountain | |
CN109325935A (en) | A kind of transmission line faultlocating method based on unmanned plane image | |
CN102654902A (en) | Contour vector feature-based embedded real-time image matching method | |
CN103400379B (en) | The detection of a kind of icing based on image and thickness method of estimation | |
CN102346015A (en) | Method for measuring icing thickness of insulator of power transmission line based on video difference analysis | |
CN107727658A (en) | Pantograph crackle on-line measuring device and method based on image procossing | |
CN111539436B (en) | Rail fastener positioning method based on straight template matching | |
CN104143192A (en) | Calibration method and device of lane departure early warning system | |
CN104599286A (en) | Optical flow based feature tracking method and device | |
CN109635737A (en) | Automobile navigation localization method is assisted based on pavement marker line visual identity | |
CN109269403A (en) | A kind of contact line conducting wire measurement of wear method and device based on machine vision | |
CN104504675A (en) | Active vision positioning method | |
CN105956619A (en) | Container lockhole coarse positioning and tracking method | |
CN104281857A (en) | Insulator detecting method based on orthogonal direction projection | |
CN108961150A (en) | Photo control point method of deploying to ensure effective monitoring and control of illegal activities automatically based on unmanned plane image | |
Wu et al. | Adjacent lane detection and lateral vehicle distance measurement using vision-based neuro-fuzzy approaches | |
CN101620672B (en) | Method for positioning and identifying three-dimensional buildings on the ground by using three-dimensional landmarks | |
Sirmacek et al. | Damaged building detection in aerial images using shadow information | |
CN110135335B (en) | Group target detection algorithm based on linear space distribution relation and projection characteristics |
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 |