CN110910443A

CN110910443A - Contact net geometric parameter real-time measuring method and device based on single monitoring camera

Info

Publication number: CN110910443A
Application number: CN201911228863.3A
Authority: CN
Inventors: 佘朝富; 宋平; 周蕾; 张楠
Original assignee: Chengdu Tang Source Electrical Ltd By Share Ltd
Current assignee: Chengdu Tang Source Electrical Ltd By Share Ltd
Priority date: 2019-12-04
Filing date: 2019-12-04
Publication date: 2020-03-24
Anticipated expiration: 2039-12-04
Also published as: CN110910443B

Abstract

The invention discloses a method and a device for measuring geometrical parameters of a contact net in real time based on a single monitoring camera, and relates to the technical field of contact net measurement. The method for measuring the geometrical parameters of the contact net, provided by the invention, can be suitable for application scenes under different illumination conditions and different background conditions, can better distinguish a lead, a carrier cable and a dropper, can reduce the frame rate and pixel requirements on a camera, and can accurately measure the geometrical parameters of the contact net.

Description

Contact net geometric parameter real-time measuring method and device based on single monitoring camera

Technical Field

The invention relates to the technical field of contact network measurement, in particular to a method and a device for measuring geometric parameters of a contact network in real time based on a single monitoring camera.

Background

The catenary is a high-voltage power transmission line for supplying power to electric locomotives, and is responsible for directly transmitting electric energy obtained from traction power transformation to trains for use. The two most critical parameters of the contact net geometric parameters are the lead height and the pull-out value of a contact net lead, and the lead height and the pull-out value are respectively defined as the vertical distance between the lead and the plane on the track and the horizontal distance between the lead and the central line of the track. According to the construction characteristics of railways in China, the geometric position of a contact network in space is easy to change, and particularly under the condition that high-speed motor train units are put into operation in large batches, each geometric parameter needs to be measured and corrected frequently. Therefore, the real state of the contact network can be reliably measured in real time, which is an important guarantee for normal operation and safe driving of the electrified railway and provides a theoretical basis for daily maintenance and overhaul of the contact network.

The current methods for detecting the geometrical parameters of the contact net mainly comprise the following methods: 1) contact measurement, such as the drop wire method, the insulation measuring rod method, and the pantograph mount sensor method. The contact measurement needs to contact the measurement equipment with the pantograph, and the method has the defects of low efficiency, large workload and low measurement precision; 2) non-contact measurements, such as ultrasound measurements, lidar methods, image detection methods. Although the method is non-contact detection and can measure the geometric parameters of the contact network in both static state and dynamic state, the method has the defects of complex installation, complex calibration, extremely high requirements on the frame rate and pixels of a camera and the like.

The national intellectual property office discloses an invention patent application with the publication number of CN110174059A and the name of 'a pantograph height and pull-out value measuring method based on monocular images' in 2019, 27.8.8.78. Due to the change of shooting conditions, the position of the pantograph in the image is changed, and the extraction of the ROI by using the fixed rectangular region is easy to cause target loss.

In addition, when the contour line and the contact line of the pantograph slide bar are analyzed by adopting an image detection method at present, a lot of interferences exist in an intercepted image, so that the contour line and/or the contact line of the pantograph slide bar are easy to have extraction errors, and the height of the pantograph slide bar and the calculation errors of a pull-out value are caused. When the geometric parameters of the contact net are measured by adopting an image detection method, how to eliminate the interference in the image so as to obtain accurate contour lines and/or contact lines is the important factor in measuring the geometric parameters of the contact net by adopting the image detection method, and the current method for measuring the geometric parameters of the contact net does not solve the problems.

Disclosure of Invention

In order to overcome the defects and shortcomings in the prior art, the invention provides a method for measuring geometrical parameters of a contact net in real time based on a single monitoring camera, and aims to solve the problems that interference images cannot be eliminated, the measurement precision is low, and errors are large in the prior art.

In order to solve the problems in the prior art, the invention is realized by the following technical scheme:

the method comprises the steps of image input, pantograph positioning, pantograph slide bar contour extraction, contact line extraction, intersection point acquisition and height and pull value calculation;

the pantograph slide bar contour extraction step specifically includes the steps of performing linear gray level enhancement on the pantograph region image reserved in the pantograph positioning step, and calculating the edge of the pantograph region image after the linear gray level enhancement by adopting a second-order difference method; calculating the length of the edge, and removing the shorter edge; and after the shorter edge is removed, connecting end points at two ends of the reserved edge into a straight line, judging an included angle between the straight line and the horizontal direction, reserving edge lines with the included angle meeting a set angle, and selecting an edge line at the uppermost end from the reserved edge lines to be used as a pantograph slide bar contour line for output. Those skilled in the art will appreciate that the shorter edge refers to an image edge having an edge length that is significantly less than the contour of the pantograph slider.

The contact line extraction step specifically includes the steps of obtaining a gradient map in the horizontal direction in the pantograph region image according to the pantograph region image reserved in the pantograph positioning step, and performing binarization processing on the gradient amplitude of the gradient map to obtain a binary map; carrying out opening operation processing on the binary image, carrying out straight line detection on the image subjected to the opening operation processing by utilizing Hough transform, and detecting all possible straight lines in the image; calculating the slope of each detected straight line in sequence, and reserving the straight lines with the slopes meeting set conditions; calculating the width of the line where each straight line is located, and screening out candidate lines meeting the set width; positioning a wire clamp area in the pantograph area image, and if a wire clamp is positioned, taking a candidate line in the center of the wire clamp area as a contact line to be output linearly; and if the line clamp cannot be positioned, sequentially calculating the distance and the slope difference between the output contact line in the image of the previous frame and the candidate line of the current frame, and selecting one candidate line with the shortest distance and/or the smallest slope difference as the contact line to be output linearly. In the preferred embodiment, the candidate lines are selected as the contact line straight line output by integrating two factors of shortest distance and minimum slope difference, so that the contact line straight line extraction accuracy under the condition that the current frame cannot be positioned to the line clamp is improved.

The pantograph positioning step specifically comprises the steps of carrying out binarization processing on an input image and extracting a foreground image; solving the area outline of the extracted foreground image; and sequentially calculating the geometric features of the outlines of the areas, judging whether the geometric features of the outlines of the areas meet a set geometric feature threshold value, if not, determining the areas as interference areas, and filtering the interference areas to only reserve pantograph areas.

In the pantograph positioning step, extracting the foreground image specifically means that each pixel point in the input image is traversed, a pixel within a range of [0, n ] is assigned as 255, and the other pixels are assigned as 0, where n is greater than or equal to 10 and less than or equal to 30.

The specific step of obtaining the region contour of the extracted foreground image is that different boundaries are endowed with different integer values, the boundaries are determined to be outer boundaries, hole boundaries and hierarchical relations of the hole boundaries, then pixels on the boundaries are marked from a starting point, and the region contour is finally formed.

The geometric features of the zone profile include at least one of zone perimeter, zone length, zone width, zone mass, and zone centroid.

The perimeter of the region refers to the number of contour points of the boundary of the region, and the length and the width of the region refer to the external rectangle of the region, and the length and the width of the region are obtained according to the external rectangle; the regional quality means that the zeroth order moment of the image is expressed as

Wherein

Represent an image in

The gray value at the point, when the image is a binary image,

the sum of pixels of a certain white area of the image is used for solving the gray quality of the white area; the centroid of the region means that the first moment of the image is expressed as

，

When the image is a binary image,

is a value of 0 or 255,

representing the accumulation of the x-coordinates of the white area on the image,

representing the accumulation of y-coordinates of white areas on the image, the centroid of the image

，

。

In a preferred embodiment, respective thresholds are set for the above 4 types of geometric features, and for a certain contour, when all the calculated geometric feature values satisfy the threshold range, the current region is retained, so that other interference regions are filtered out, and finally only the pantograph region is retained. By integrating the 4 types of geometric characteristics provided by the application, the accuracy of the pantograph region positioning is obviously improved.

And performing linear gray scale enhancement on the reserved pantograph region image, specifically, converting the gray scales of all points in the image according to a linear gray scale conversion function. In particular, images

Representing the gray scale range before transformation is

The transformed image is

Then, the gray scale linear transformation function expression is:

(ii) a Transformed image

To expand the gray scale range to

。

The method for obtaining the gradient map in the horizontal direction in the pantograph region image comprises the following specific steps:

pixel point in image

The gradient of (d) is:

（1）。

the binarization processing of the gradient amplitude of the gradient map specifically includes setting a threshold value and converting G_x（x,y) The pixels larger than the threshold are set to 255, and the remaining pixels are set to 0.

And calculating the width of the line where each straight line is located, specifically, taking any point of the straight line as a starting point, extracting the gray value gray of the straight line, searching pixel points leftwards and rightwards along the x direction, if the gray value of the point is equal to the gray, adding 1 to the width of the line, and otherwise, stopping searching.

In the intersection point obtaining step, the pantograph slide bar contour line output in the pantograph slide bar contour line extracting step is received, a coordinate point set of the pantograph slide bar contour line is created according to the received pantograph slide bar contour line, the maximum value and the minimum value of the abscissa are found out from the coordinate point set, and the median of the abscissa is obtained according to the maximum value and the minimum value of the abscissa; sorting coordinate points in the coordinate point set of the pantograph slide bar contour line from small to large by taking the abscissa as a reference, and finding out the coordinate point of which the abscissa is equal to the solved median of the abscissa

In terms of coordinate points

A plurality of coordinate points are respectively expanded to the left and right of the adjacent central point, and are intercepted to form a new pantograph slider contour line coordinate point set; and performing linear fitting on the new coordinate point set of the pantograph slider contour line by using a least square method to obtain a linear equation of the coordinate point set, and solving the intersection point according to the linear equation of the contact line extracted in the contact line extraction step and the linear method of the pantograph slider contour line obtained after linear fitting.

Specifically, two end points p1, p2 on the contact line are known, and the linear expression of the contact line is as follows:

A1X+B1Y+C1=0 （2）；

two end points p3, p4 on the contour line of the pantograph slide bar obtained after straight line fitting, and the straight line expression is as follows:

A2X+B2Y+C2=0 （3）；

simultaneous equations (5) and (6), and let B1= -1, B2= -1, one can obtain:

X=(C1+Y)/A1 （4）；

Y=(A2C1-A1C2)/(A2-A1) （5）；

according to the known coordinate points, the values of equation coefficients A1, B1, C1, A2, B2 and C2 are obtained through solving, and the intersection point is further obtained according to the obtained equation coefficients.

In the step of obtaining the height guiding and pulling-out value, the camera calibration condition is that the coordinate of the center point x of the pantograph is superposed with the coordinate of the center point of the chessboard grid calibration plate, and the coordinate of the y is not required; the world coordinate system is established in the center of the chessboard, the horizontal right direction is the positive direction of x, and the vertical upward direction is the positive direction of y; the camera calibration means that intersection point image coordinates are converted into established world coordinates to obtain an actual physical position of an intersection point, and if coordinates obtained by converting an intersection point P into a world coordinate system are P (x, y), an obtained derivative height H and a pull-out value S are as follows:

H=L+Py （6）；

s = Px (7); wherein L represents the distance of the center point of the checkerboard from the rail surface.

The invention also provides a device for measuring the geometric parameters of the overhead line system in real time based on the single monitoring camera, which comprises an image input module, a pantograph positioning module, a pantograph slider outline extraction module, a contact line extraction module, an intersection point acquisition module and a lead-height and pull-out value calculation module; the image input module is used for receiving the image transmitted by the single monitoring camera and transmitting the received image to the pantograph positioning module; the pantograph positioning module is used for positioning a pantograph area image in the image received by the image input module and transmitting the reserved pantograph area image to the pantograph slide bar outline extraction module and the contact line extraction module;

the pantograph slide bar outline extraction module is used for performing linear gray level enhancement on a pantograph region image reserved in the pantograph positioning module and calculating the edge of the pantograph region image subjected to linear gray level enhancement by adopting a second-order difference method; calculating the length of the edge, and removing the shorter edge; after removing the shorter edge, connecting end points at two ends of the reserved edge into a straight line, judging an included angle between the straight line and the horizontal direction, reserving edge lines with the included angle meeting a set angle, selecting an edge line at the uppermost end from the reserved edge lines as a pantograph slide bar contour line, and outputting the edge line to an intersection point solving module;

the contact line extraction module is used for receiving the pantograph region image transmitted by the pantograph positioning module, solving a gradient map in the horizontal direction in the pantograph region image according to the pantograph region image, and performing binarization processing on a gradient amplitude value of the gradient map to obtain a binary map; carrying out opening operation processing on the binary image, carrying out straight line detection on the image subjected to the opening operation processing by utilizing Hough transform, and detecting all possible straight lines in the image; calculating the slope of each detected straight line in sequence, and reserving the straight lines with the slopes meeting set conditions; calculating the width of the line where each straight line is located, and screening out candidate lines meeting the set width; positioning a wire clamp area in the pantograph area image, and if a wire clamp is positioned, taking a candidate line in the center of the wire clamp area as a contact line to be output linearly; if the line clamp can not be positioned, the distance and the slope difference between the output contact line in the image of the previous frame and the candidate line of the current frame are sequentially calculated, and one candidate line with the shortest distance and/or the smallest slope difference is selected as the contact line straight line to be output to the intersection point solving module.

The pantograph positioning module is used for carrying out binarization processing on an input image and extracting a foreground image; solving the area outline of the extracted foreground image; and sequentially calculating the geometric features of the outlines of the areas, judging whether the geometric features of the outlines of the areas meet a set geometric feature threshold value, if not, determining the areas as interference areas, and filtering the interference areas to only reserve pantograph areas.

In the pantograph positioning module, extracting the foreground image specifically means traversing each pixel point in the input image, assigning pixels within a range of [0, n ], 10 ≦ n ≦ 30, to 255, and assigning the remaining pixels to 0.

Wherein

Represent an image in

The gray value at the point, when the image is a binary image,

，

When the image is a binary image,

is a value of 0 or 255,

，

。

Representing the gray scale range before transformation is

After transformation into

Then, the gray scale linear transformation function expression is:

(ii) a Transformed image

To expand the gray scale range to

。

The method for obtaining the horizontal gradient map in the pantograph region image comprises the following specific steps:

pixel point in image

The gradient of (d) is:

（1）。

The intersection point calculating module receives the pantograph slide bar contour line output in the pantograph slide bar contour line extracting step, creates a pantograph slide bar contour line coordinate point set according to the received pantograph slide bar contour line, finds out an abscissa maximum value and an abscissa minimum value from the coordinate point set, and calculates an abscissa median value according to the abscissa maximum value and the abscissa minimum value; sorting coordinate points in the coordinate point set of the pantograph slide bar contour line from small to large by taking the abscissa as a reference, and finding out the coordinate point of which the abscissa is equal to the solved median of the abscissa

In terms of coordinate points

A1X+B1Y+C1=0 （2）；

A2X+B2Y+C2=0 （3）；

simultaneous equations (5) and (6), and let B1= -1, B2= -1, one can obtain:

X=(C1+Y)/A1 （4）；

Y=(A2C1-A1C2)/(A2-A1) （5）；

In the height leading and pulling value obtaining module, the camera calibration condition is that the coordinate of the center point x of the pantograph is superposed with the coordinate of the center point of the chessboard grid calibration plate, and the coordinate of the y is not required; the world coordinate system is established in the center of the chessboard, the horizontal right direction is the positive direction of x, and the vertical upward direction is the positive direction of y; the camera calibration means that intersection point image coordinates are converted into established world coordinates to obtain an actual physical position of an intersection point, and if coordinates obtained by converting an intersection point P into a world coordinate system are P (x, y), an obtained derivative height H and a pull-out value S are as follows:

H=L+Py （6）；

Compared with the prior art, the beneficial technical effects brought by the invention are as follows:

1. in the invention, further improvements are respectively provided for a pantograph positioning step, a pantograph slide bar outline extraction step and a contact line extraction step in the existing method for measuring the geometrical parameters of the overhead line system of the single monitoring camera in real time. In the pantograph positioning step, foreground image extraction, contour extraction and contour screening are adopted, so that the pantograph region is accurately positioned.

2. The pantograph slider outline extraction step comprises image enhancement, namely, the gray levels of all points in an image are converted according to a linear gray level conversion function, each pixel in the image is linearly expanded, and the image quality is effectively improved; the method also comprises line detection, and the image edge is calculated by adopting a second-order difference method, so that the calculation precision is higher. The method also comprises line screening, wherein the contour lines are screened, so that one contour line is determined to be output as the contour line of the pantograph slider, the accuracy of the contour line of the pantograph slider is ensured, and the accuracy of subsequent height guidance and pull-out value calculation is improved.

3. The contact line extraction step comprises gradient calculation, and aims to overcome the influence of illumination, ensure that the integral gray difference of the image under different light bars is larger, and the gradient characteristics in a specific direction are relatively stable. Because the gradient amplitude represents the intensity of pixel mutation, and the edge of the wire is an area with stronger pixel mutation, the gradient amplitude is subjected to binarization processing, and a lot of noise exists after the binarization processing of the gradient image, so that the binary image is subjected to open operation processing, the contour is smoothed, narrow connection is broken, and a small salient part is removed. And then, carrying out straight line detection on the morphologically processed image by using Hough transform, detecting all possible straight lines in the image, wherein a plurality of initial straight lines are extracted, such as contact lines, messenger lines, suspenders, inclined pull lines and the like, and interference straight lines (messenger lines, suspenders, inclined pull lines and the like) need to be removed. In the invention, a slope calculation mode is adopted, so that a part of unqualified straight lines are filtered, then a part of unqualified straight lines are filtered according to the line width, and then a line clamp area is positioned, if a line clamp is positioned, because a real contact line is clamped by the line clamp, a candidate line at the center of the line clamp area is taken as a contact line straight line to be output; if the line clamp can not be positioned, the distance and the slope difference between the output contact line and a plurality of wires of the current frame in the image of the current frame are calculated in sequence, and one line with the shortest distance and the smallest slope difference is selected as the linear output of the contact line (if the first input image is adopted, the real contact line is marked manually).

4. The pantograph contour lines are screened through the edge line length, the edge line included angle and the position characteristics of the edge lines, the extraction precision of the pantograph contour lines is further improved, and the pantograph contour lines are guaranteed to be extracted correctly.

5. The method is insensitive to the light source, can be suitable for calculating the leading-out value and the pull-out value of the image under different illumination conditions, and can be suitable for both day and night. The method has strong field applicability, the algorithm parameters of the method are convenient to modify, no professional is needed, other non-professionals can modify the related parameters, and the method can be suitable for calculating the leading-up and the pull-out values of different background images under different field conditions. The invention has high detection accuracy and low false detection rate and missed detection rate.

Drawings

FIG. 1 is a flow chart of the measurement of geometrical parameters of the overhead line system of the present invention;

FIG. 2 is an original photograph taken by the camera of the present invention;

FIG. 3 is a diagram of pantograph positioning according to the present invention;

FIG. 4 is a diagram of an outline extraction of a pantograph in accordance with the present invention;

FIG. 5 is a schematic diagram of a multi-interference line of the present invention;

FIG. 6 is a previous frame of image in the straight line filtering step according to the present invention;

FIG. 7 is a current frame image in the straight line filtering step of the present invention;

FIG. 8 is a cross-point output plot of the present invention.

Detailed Description

The technical scheme of the invention is further elaborated in the following by combining the drawings in the specification.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as methods, apparatus, computer program products. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

Example 1

Referring to fig. 1-8 of the specification, this embodiment discloses: a contact net geometric parameter real-time measurement method based on a single monitoring camera can accurately extract a pantograph slider contour line, can well distinguish a lead, a carrier cable and a dropper, accurately extract a contact line, improve measurement precision and reduce errors. Specifically, the method comprises the following steps:

the method comprises an image input step, a pantograph positioning step, a pantograph slide bar outline extraction step, a contact line extraction step, an intersection point obtaining step and a height and pull value calculation step, wherein the pantograph slide bar outline extraction step specifically refers to the steps of performing linear gray level enhancement on a pantograph region image reserved in the pantograph positioning step and calculating the edge of the pantograph region image after linear gray level enhancement by adopting a second-order difference method; calculating the length of the edge, and removing the shorter edge; after removing the shorter edge, connecting end points at two ends of the reserved edge into a straight line, judging an included angle between the straight line and the horizontal direction, reserving edge lines with the included angle meeting a set angle, and selecting an edge line at the uppermost end from the reserved edge lines as a pantograph slide bar contour line for output;

the contact line extraction step specifically includes the steps of obtaining a horizontal gradient map in the pantograph region image according to the pantograph region image reserved in the pantograph positioning step, and performing binarization processing on the gradient amplitude of the gradient map to obtain a binary map; carrying out opening operation processing on the binary image, carrying out straight line detection on the image subjected to the opening operation processing by utilizing Hough transform, and detecting all possible straight lines in the image; calculating the slope of each detected straight line in sequence, and reserving the straight lines with the slopes meeting set conditions; calculating the width of the line where each straight line is located, and screening out candidate lines meeting the set width; positioning a wire clamp area in the pantograph area image, and if a wire clamp is positioned, taking a candidate line in the center of the wire clamp area as a contact line to be output linearly; and if the line clamp cannot be positioned, calculating the distance and the slope difference between the output contact line in the image of the previous frame and the candidate line of the current frame in sequence, and selecting one candidate line with the shortest distance and the smallest slope difference as the contact line to be output linearly.

Inputting an acquired image of a single monitoring camera, wherein the image format is JPG coding, the image needs to be decoded into an original gray image, and the image input size is 2048 x 1536; the pantograph positioning step specifically comprises the steps of carrying out binarization processing on an input image and extracting a foreground image; solving the area outline of the extracted foreground image; and sequentially calculating the geometric features of the outlines of the areas, judging whether the geometric features of the outlines of the areas meet a set geometric feature threshold value, if not, determining the areas as interference areas, and filtering the interference areas to only reserve pantograph areas.

In the pantograph positioning step, extracting the foreground image specifically means traversing each pixel point in the input image, assigning the pixels within the range of [0, n ], where n is greater than or equal to 10 and less than or equal to 30 to 255, and assigning the remaining pixels to 0.

Specifically, a foreground image is input to

Represented as pixel values of the image, each line scan is terminated with two cases: (1)

at this time

Is the starting point of the outer boundary; (2)

at this time

Is the starting point of the hole boundary; then, starting from the starting point, pixels on the boundary are marked, and finally, the region outline is formed.

The geometric characteristics of the region outline comprise region perimeter, region length, region width, region mass and region centroid.

Wherein

Represent an image in

The gray value at the point, when the image is a binary image,

is the sum of pixels of a white area of the image, which is usedSolving the gray quality of the white area; the centroid of the region means that the first moment of the image is expressed as

，

When the image is a binary image,

is a value of 0 or 255,

，

。

Representing the gray scale range before transformation is

After transformation into

Then, the gray scale linear transformation function expression is:

(ii) a Transformed image

To expand the gray scale range to

。

The method for calculating the edge of the pantograph region image after linear gray scale enhancement by adopting a second-order difference method specifically means that under the condition of a continuous function, a maximum value or a minimum value in a first-order differential diagram represents the edge, a zero crossing point between the maximum value and the minimum value in the second-order differential diagram represents the edge, and then a first-order difference represents the edge

(ii) a The second order difference is expressed as:

；

it can be derived from this that,

extracting the coefficients [1, -2,1 ] thereof](ii) a In the case of two dimensions, the same can be said:

，

extracting coefficients of the above functions, expressed in the form of a template

And performing convolution operation on the obtained two-dimensional template and the image to obtain the image edge.

In the step of extracting the outline of the pantograph slide bar, a rectangular coordinate system is established in a reserved pantograph area, a Y axis is represented in a top-down direction, an X axis is represented in a left-to-right direction, the edge length is calculated according to the edge of an image calculated by a second-order difference method, and a shorter edge is removed; sorting the coordinate points of each edge by taking the X axis as a reference; selecting end points at two ends of the edge line to be connected into a straight line, calculating an included angle between the straight line and the x axis, and leaving the edge line which accords with a set angle; and calculating the distance between a straight line formed by connecting two end points of the left edge line and a straight line with y =0, and selecting a contour line closest to the straight line as a pantograph slider contour line for output.

Or in the step of extracting the outline of the pantograph slide bar, establishing a rectangular coordinate system by using a reserved pantograph region, representing a Y axis by using a direction from bottom to top, representing an X axis by using a direction from left to right, calculating the edge length according to the edge of the image calculated by a second-order difference method, and removing a shorter edge; sorting the coordinate points of each edge by taking the X axis as a reference; selecting end points at two ends of the edge line to be connected into a straight line, calculating an included angle between the straight line and the x axis, and leaving the edge line which accords with a set angle; and calculating the distance between a straight line formed by connecting two end points of the left edge line and a straight line with y =0, and selecting a contour line with the farthest distance as a pantograph slide bar contour line for output.

pixel point in image

The gradient of (d) is:

（1）

the binarization processing of the horizontal gradient amplitude of the gradient map specifically means setting a threshold value to be used for binarization processingG _x(x,y) The pixels larger than the threshold are set to 255, and the remaining pixels are set to 0.

In the intersection point obtaining step, the pantograph slide bar contour line output in the pantograph slide bar contour line extracting step is received, and a pantograph slide bar wheel is created according to the received pantograph slide bar contour lineA coordinate point set of the profile line is used for finding out the maximum value and the minimum value of the abscissa from the coordinate point set, and the median value of the abscissa is obtained according to the maximum value and the minimum value of the abscissa; sorting coordinate points in the coordinate point set of the pantograph slide bar contour line from small to large by taking the abscissa as a reference, and finding out the coordinate point of which the abscissa is equal to the solved median of the abscissa

In terms of coordinate points

A1X+B1Y+C1=0 （2）；

A2X+B2Y+C2=0 （3）；

simultaneous equations (5) and (6), and let B1= -1, B2= -1, one can obtain:

X=(C1+Y)/A1 （4）；

Y=(A2C1-A1C2)/(A2-A1) （5）；

H=L+Py （6）；

Example 2

Referring to fig. 1-8 of the specification, this embodiment discloses as another preferred embodiment of the present invention: a device for measuring geometrical parameters of a contact network in real time based on a single monitoring camera comprises an image input module, a pantograph positioning module, a pantograph slide bar outline extraction module, a contact line extraction module, an intersection point acquisition module and a height and pull value calculation module; the image input module is used for receiving the image transmitted by the single monitoring camera and transmitting the received image to the pantograph positioning module; the pantograph positioning module is used for positioning a pantograph area image in the image received by the image input module and transmitting the reserved pantograph area image to the pantograph slide bar outline extraction module and the contact line extraction module;

the contact line extraction module is used for receiving the pantograph region image transmitted by the pantograph positioning module, solving a horizontal gradient map in the pantograph region image according to the pantograph region image, and performing binarization processing on a gradient amplitude value of the gradient map to obtain a binary map; carrying out opening operation processing on the binary image, carrying out straight line detection on the image subjected to the opening operation processing by utilizing Hough transform, and detecting all possible straight lines in the image; calculating the slope of each detected straight line in sequence, and reserving the straight lines with the slopes meeting set conditions; calculating the width of the line where each straight line is located, and screening out candidate lines meeting the set width; positioning a wire clamp area in the pantograph area image, and if a wire clamp is positioned, taking a candidate line in the center of the wire clamp area as a contact line to be output linearly; if the line clamp can not be positioned, the distance and the slope difference between the output contact line in the image of the previous frame and the candidate line of the current frame are calculated in sequence, and one candidate line with the shortest distance and the smallest slope difference is selected as the contact line to be output to the intersection point solving module.

The image input module inputs the collected images of the single monitoring camera, the image format is JPG coding, the images need to be decoded into original gray images, and the image input size is 2048 x 1536; the pantograph positioning module is used for carrying out binarization processing on an input image and extracting a foreground image; solving the area outline of the extracted foreground image; and sequentially calculating the geometric features of the outlines of the areas, judging whether the geometric features of the outlines of the areas meet a set geometric feature threshold value, if not, determining the areas as interference areas, and filtering the interference areas to only reserve pantograph areas.

Specifically, a foreground image is input to

at this time

Is the starting point of the outer boundary; (2)

at this time

The geometric features of the zone profile include zone perimeter, zone length, zone width, zone mass, and zone centroid.

Wherein

Represent an image in

The gray value at the point, when the image is a binary image,

，

When the image is a binary image,

is a value of 0 or 255,

，

。

Representing the gray scale range before transformation is

After transformation into

Then, the gray scale linear transformation function expression is:

(ii) a Transformed image

To expand the gray scale range to

。

(ii) a The second order difference is expressed as:

；

it can be derived from this that,

，

The pantograph slider outline extraction module establishes a rectangular coordinate system according to a reserved pantograph region, represents a Y axis in a top-down direction, represents an X axis in a left-to-right direction, calculates the edge length according to the edge of the image calculated by a second-order difference method, and removes a shorter edge; sorting the coordinate points of each edge by taking the X axis as a reference; selecting end points at two ends of the edge line to be connected into a straight line, calculating an included angle between the straight line and the x axis, and leaving the edge line which accords with a set angle; and calculating the distance between a straight line formed by connecting two end points of the left edge line and a straight line with y =0, and selecting a contour line closest to the straight line as a pantograph slider contour line for output.

Or the pantograph slide bar outline extraction module establishes a rectangular coordinate system by using a reserved pantograph region, represents a Y axis by using a direction from bottom to top, represents an X axis by using a direction from left to right, calculates the edge length according to the edge of the image calculated by a second-order difference method, and removes a shorter edge; sorting the coordinate points of each edge by taking the X axis as a reference; selecting end points at two ends of the edge line to be connected into a straight line, calculating an included angle between the straight line and the x axis, and leaving the edge line which accords with a set angle; and calculating the distance between a straight line formed by connecting two end points of the left edge line and a straight line with y =0, and selecting a contour line with the farthest distance as a pantograph slide bar contour line for output.

pixel point in image

The gradient of (d) is:

（1）

the binarization processing of the gradient amplitude of the gradient map specifically means setting a threshold value to be used for binarization processingG _x(x,y) The pixels larger than the threshold are set to 255, and the remaining pixels are set to 0.

The intersection point calculating module receives the pantograph slide bar contour line output in the pantograph slide bar contour line extracting step, creates a pantograph slide bar contour line coordinate point set according to the received pantograph slide bar contour line, finds out an abscissa maximum value and an abscissa minimum value from the coordinate point set, and calculates an abscissa median value according to the abscissa maximum value and the abscissa minimum value; in the abscissaSequencing the coordinate points in the coordinate point set of the pantograph slide bar contour line from small to large as a reference, and finding out the coordinate point of which the abscissa is equal to the solved median of the abscissas

In terms of coordinate points

A1X+B1Y+C1=0 （2）；

A2X+B2Y+C2=0 （3）；

simultaneous equations (5) and (6), and let B1= -1, B2= -1, one can obtain:

X=(C1+Y)/A1 （4）；

Y=(A2C1-A1C2)/(A2-A1) （5）；

H=L+Py （6）；

Example 3

As another preferred embodiment of the present invention, the present embodiment discloses a pantograph profile extraction method based on a single monitoring camera, which can accurately extract a pantograph slider profile and improve measurement accuracy.

A pantograph contour extraction method based on a single monitoring camera comprises an image input step, a pantograph positioning step and a pantograph slide bar contour extraction step, wherein the pantograph slide bar contour extraction step specifically refers to the steps of performing linear gray level enhancement on a pantograph region image reserved in the pantograph positioning step, and calculating the edge of the pantograph region image after the linear gray level enhancement by adopting a second-order difference method; calculating the length of the edge, and removing the shorter edge; and after the shorter edge is removed, connecting end points at two ends of the reserved edge into a straight line, judging an included angle between the straight line and the horizontal direction, reserving edge lines with the included angle meeting a set angle, and selecting an edge line at the uppermost end from the reserved edge lines to be used as a pantograph slide bar contour line for output.

Specifically, a foreground image is input to

at this time

Is the starting point of the outer boundary; (2)

at this time

Wherein

Represent an image in

The gray value at the point, when the image is a binary image,

，

When the image is a binary image,

is a value of 0 or 255,

，

。

Representing the gray scale range before transformation is

After transformation into

Then, the gray scale linear transformation function expression is:

(ii) a Transformed image

To expand the gray scale range to

。

(ii) a The second order difference is expressed as:

；

it can be derived from this that,

，

Performing convolution operation on the obtained two-dimensional template and the image to obtain an image edgeA rim.

Example 4

As another preferred embodiment of the present invention, the present embodiment discloses: a pantograph contour extraction device based on a single monitoring camera comprises an image input module, a pantograph positioning module and a pantograph slider contour extraction module, wherein the image input module is used for receiving a pantograph image shot by the single monitoring camera and transmitting the received image to the pantograph positioning module; the pantograph slide bar outline extraction module performs linear gray level enhancement on the pantograph region image transmitted by the pantograph positioning module, and calculates the edge of the pantograph region image after linear gray level enhancement by adopting a second-order difference method; calculating the length of the edge, and removing the shorter edge; and after the shorter edge is removed, connecting end points at two ends of the reserved edge into a straight line, judging an included angle between the straight line and the horizontal direction, reserving edge lines with the included angle meeting a set angle, and selecting an edge line at the uppermost end from the reserved edge lines to be used as a pantograph slide bar contour line for output.

The image input module receives an acquired image of the single monitoring camera, the image format is JPG coding, the image is decoded into an original gray image, and the image output size is 2048 x 1536; the method comprises the steps of outputting an image to a pantograph positioning module, wherein the pantograph positioning module is used for carrying out binarization processing on the input image and extracting a foreground image; solving the area outline of the extracted foreground image; and sequentially calculating the geometric features of the outlines of the areas, judging whether the geometric features of the outlines of the areas meet a set geometric feature threshold value, if not, determining the areas as interference areas, and filtering the interference areas to only reserve pantograph areas.

The pantograph positioning module extracts the foreground image, specifically, traversing each pixel point in the input image, assigning a pixel within a range of [0, n ], 10 ≦ n ≦ 30, to 255, and assigning the remaining pixels to 0.

Specifically, a foreground image is input to

at this time

Is the starting point of the outer boundary; (2)

at this time

Wherein

Represent an image in

The gray value at the point, when the image is a binary image,

，

When the image is a binary image,

is a value of 0 or 255,

，

。

Representing the gray scale range before transformation is

After transformation into

Then, the gray scale linear transformation function expression is:

(ii) a Transformed image

To expand the gray scale range to

。

(ii) a The second order difference is expressed as:

；

it can be derived from this that,

，

Furthermore, the pantograph slider outline extraction module establishes a rectangular coordinate system according to a reserved pantograph region, represents a Y axis in a top-down direction, represents an X axis in a left-to-right direction, calculates the edge length according to the edge of the image calculated by the second-order difference method, and removes a shorter edge; sorting the coordinate points of each edge by taking the X axis as a reference; selecting end points at two ends of the edge line to be connected into a straight line, calculating an included angle between the straight line and the x axis, and leaving the edge line which accords with a set angle; and calculating the distance between a straight line formed by connecting two end points of the left edge line and a straight line with y =0, and selecting a contour line closest to the straight line as a pantograph slider contour line for output.

Example 5

As another preferred embodiment of the present invention, this embodiment discloses a contact line extraction method based on a single monitoring camera, and the present embodiment provides a contact line extraction method based on a single monitoring camera, which aims to solve the problems in the prior art that the interference of a contact line in an image cannot be eliminated, the contact line cannot be accurately extracted, the measurement accuracy is low, and the error is large. The contact line extraction method based on the single monitoring camera provided by the embodiment can well distinguish the lead, the carrier cable and the dropper, accurately extract the contact line, improve the measurement precision and reduce the error.

A contact line extraction method based on a single monitoring camera comprises an image input step, a pantograph positioning step and a contact line extraction step, wherein the contact line extraction step specifically comprises the steps of obtaining a horizontal gradient map in a pantograph area image according to the pantograph area image reserved in the pantograph positioning step, and carrying out binarization processing on the gradient amplitude of the gradient map to obtain a binary map; carrying out opening operation processing on the binary image, carrying out straight line detection on the image subjected to the opening operation processing by utilizing Hough transform, and detecting all possible straight lines in the image; calculating the slope of each detected straight line in sequence, and reserving the straight lines with the slopes meeting set conditions; calculating the width of the line where each straight line is located, and screening out candidate lines meeting the set width; positioning a wire clamp area in the pantograph area image, and if a wire clamp is positioned, taking a candidate line in the center of the wire clamp area as a contact line to be output linearly; and if the line clamp cannot be positioned, calculating the distance and the slope difference between the output contact line in the image of the previous frame and the candidate line of the current frame in sequence, and selecting one candidate line with the shortest distance and the smallest slope difference as the contact line to be output linearly.

And an image input step, wherein the acquired image of the single monitoring camera is input, the image format is JPG coding, the image needs to be decoded into an original gray image, and the image input size is 2048 × 1536.

Specifically, a foreground image is input to

at this time

Is the starting point of the outer boundary; (2)

at this time

Wherein

Represent an image in

The gray value at the point, when the image is a binary image,

，

When the image is a binary image,

is a value of 0 or 255,

，

。

pixel point in image

The gradient of (d) is:

（1）；

the binarization processing of the gradient amplitude of the gradient map specifically means setting a threshold value to be used for binarization processing

The pixels larger than the threshold are set to 255, and the remaining pixels are set to 0.

Example 6

As another preferred embodiment of the present invention, the embodiment discloses a contact line extraction device based on a single monitoring camera, including an image input module, a pantograph positioning module and a contact line extraction module, where the image input module is configured to receive a pantograph image captured by the single monitoring camera and transmit the received image to the pantograph positioning module, the pantograph positioning module processes the received image, reserves a pantograph region image, and transmits the reserved pantograph region image to the contact line extraction module, and the contact line extraction module, according to the pantograph region image reserved in the pantograph positioning module, obtains a horizontal gradient map in the pantograph region image, and performs binarization processing on a gradient amplitude of the gradient map to obtain a binary map; carrying out opening operation processing on the binary image, carrying out straight line detection on the image subjected to the opening operation processing by utilizing Hough transform, and detecting all possible straight lines in the image; calculating the slope of each detected straight line in sequence, and reserving the straight lines with the slopes meeting set conditions; calculating the width of the line where each straight line is located, and screening out candidate lines meeting the set width; positioning a wire clamp area in the pantograph area image, and if a wire clamp is positioned, taking a candidate line in the center of the wire clamp area as a contact line to be output linearly; and if the line clamp cannot be positioned, calculating the distance and the slope difference between the output contact line in the image of the previous frame and the candidate line of the current frame in sequence, and selecting one candidate line with the shortest distance and the smallest slope difference as the contact line to be output linearly.

The image input module inputs the collected images of the single monitoring camera, the image format is JPG coding, the images need to be decoded into original gray images, and the image input size is 2048 x 1536.

Specifically, a foreground image is input to

at this time

Is the starting point of the outer boundary; (2)

at this time

Wherein

Represent an image in

The gray value at the point, when the image is a binary image,

，

When the image is a binary image,

is a value of 0 or 255,

，

。

pixel point in image

The gradient of (d) is:

（1）。

Example 7

In order to achieve the above object, according to another aspect of the present application, there is also provided a computer-readable storage medium storing a computer program, which when executed in a computer processor, implements the steps in the above method for measuring geometrical parameters of a catenary in real time based on a single monitoring camera.

Example 8

In order to achieve the above object, according to another aspect of the present application, there is also provided a computer-readable storage medium storing a computer program which, when executed in a computer processor, implements the steps in the above single-monitored-camera-based pantograph contour extraction method.

Example 9

In order to achieve the above object, according to another aspect of the present application, there is also provided a computer-readable storage medium storing a computer program which, when executed in a computer processor, implements the steps in the above-mentioned single monitoring camera-based contact line extraction method.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard disk (Hard disk Drive, abbreviated as HDD) or a Solid State Drive (SSD), etc.; the storage medium may also comprise a combination of memories of the kind described above. It will be apparent to those skilled in the art that the modules or steps of the present invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and they may alternatively be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, or fabricated separately as individual integrated circuit modules, or fabricated as a single integrated circuit module from multiple modules or steps. Thus, the present invention is not limited to any specific combination of hardware and software.

Example 10

In order to achieve the above object, according to another aspect of the present application, there is also provided a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the steps in the above method for measuring geometric parameters of a catenary in real time based on a single monitoring camera.

Example 11

In order to achieve the above object, according to another aspect of the present application, there is also provided a computer device including a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps in the pantograph profile extraction method based on a single monitoring camera when executing the computer program.

Example 12

In order to achieve the above object, according to another aspect of the present application, there is also provided a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps in the above contact line extraction method based on a single monitoring camera when executing the computer program.

In the above embodiments 10-12, the processor may be a Central Processing Unit (CPU). The Processor may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, or a combination thereof.

The memory, which is a non-transitory computer readable storage medium, may be used to store non-transitory software programs, non-transitory computer executable programs, and units, such as the corresponding program units in the above-described method embodiments of the present invention. The processor executes various functional applications of the processor and the processing of the work data by executing the non-transitory software programs, instructions and modules stored in the memory, that is, the method in the above method embodiment is realized.

The memory may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created by the processor, and the like. Further, the memory may include high speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory optionally includes memory located remotely from the processor, and such remote memory may be coupled to the processor via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The one or more units are stored in the memory and when executed by the processor perform the methods of embodiments 1, 3, 5 above.

The specific details of the computer device may be understood by referring to the corresponding related descriptions and effects in the above embodiments, and are not described herein again.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. The method for measuring the geometrical parameters of the contact network in real time based on the single monitoring camera comprises an image input step, a pantograph positioning step, a pantograph slide bar outline extraction step, a contact line extraction step, an intersection point obtaining step and a height and pull value calculation step, and is characterized in that:

the pantograph slide bar contour extraction step specifically comprises the following steps: performing linear gray scale enhancement on the pantograph region image reserved in the pantograph positioning step, and calculating the edge of the pantograph region image after the linear gray scale enhancement by adopting a second-order difference method; calculating the length of the edge, and removing the shorter edge; after removing the shorter edge, connecting end points at two ends of the reserved edge into a straight line, judging an included angle between the straight line and the horizontal direction, reserving edge lines with the included angle meeting a set angle, and selecting an edge line at the uppermost end from the reserved edge lines as a pantograph slide bar contour line for output;

the contact line extraction step specifically comprises the following steps: according to the pantograph region image reserved in the pantograph positioning step, obtaining a gradient map in the horizontal direction in the pantograph region image, and performing binarization processing on the gradient amplitude of the gradient map to obtain a binary map; carrying out opening operation processing on the binary image, carrying out straight line detection on the image subjected to the opening operation processing by utilizing Hough transform, and detecting all possible straight lines in the image; calculating the slope of each detected straight line in sequence, and reserving the straight lines with the slopes meeting set conditions; calculating the width of the line where each straight line is located, and screening out candidate lines meeting the set width; positioning a wire clamp area in the pantograph area image, and if a wire clamp is positioned, taking a candidate line in the center of the wire clamp area as a contact line to be output linearly; and if the line clamp cannot be positioned, sequentially calculating the distance and the slope difference between the output contact line in the image of the previous frame and the candidate line of the current frame, and selecting one candidate line with the shortest distance and/or the smallest slope difference as the contact line to be output linearly.

2. The method for measuring the geometric parameters of the overhead line system based on the single monitoring camera in real time as claimed in claim 1, wherein the method comprises the following steps: the pantograph positioning step specifically comprises the steps of carrying out binarization processing on an input image and extracting a foreground image; solving the area outline of the extracted foreground image; and sequentially calculating the geometric features of the outlines of the areas, judging whether the geometric features of the outlines of the areas meet a set geometric feature threshold value, if not, determining the areas as interference areas, and filtering the interference areas to only reserve pantograph areas.

3. The real-time measurement method for the geometric parameters of the overhead line system based on the single monitoring camera as claimed in claim 2, is characterized in that: the specific step of obtaining the region contour of the extracted foreground image is that different boundaries are endowed with different integer values, the boundaries are determined to be outer boundaries, hole boundaries and hierarchical relations of the hole boundaries, then pixels on the boundaries are marked from a starting point, and the region contour is finally formed.

4. The real-time measurement method for the geometric parameters of the overhead line system based on the single monitoring camera as claimed in claim 2, is characterized in that: the geometric features of the zone profile include at least one of zone perimeter, zone length, zone width, zone mass, and zone centroid.

5. The method for measuring the geometric parameters of the overhead line system based on the single monitoring camera in real time as claimed in claim 1, wherein the method comprises the following steps: and calculating the width of the line where each straight line is located, specifically, taking any point of the straight line as a starting point, extracting the gray value gray of the straight line, searching pixel points leftwards and rightwards along the x direction, if the gray value of the point is equal to the gray, adding 1 to the width of the line, and otherwise, stopping searching.

6. The method for measuring the geometric parameters of the overhead line system based on the single monitoring camera in real time as claimed in claim 1, wherein the method comprises the following steps: in the intersection point obtaining step, the pantograph slide bar contour line output in the pantograph slide bar contour line extracting step is received, a coordinate point set of the pantograph slide bar contour line is created according to the received pantograph slide bar contour line, the maximum value and the minimum value of the abscissa are found out from the coordinate point set, and the median of the abscissa is obtained according to the maximum value and the minimum value of the abscissa; sorting coordinate points in the coordinate point set of the pantograph slide bar contour line from small to large by taking the abscissa as a reference, and finding out the coordinate point of which the abscissa is equal to the solved median of the abscissa

In terms of coordinate points

7. The method for measuring the geometric parameters of the overhead line system based on the single monitoring camera in real time as claimed in claim 1, wherein the method comprises the following steps: in the step of obtaining the lead-height and the pull-out value, the intersection point P acquired in the image is converted into a coordinate P (x, y) of a world coordinate system according to a camera calibration result, and the obtained lead-height H and the pull-out value S are as follows:

H=L+Py；

s = Px; the camera calibration condition is to ensure that the x coordinate of the center point of the bow is coincident with the coordinate of the center point of the chessboard calibration plate, the y coordinate does not make a requirement, a world coordinate system is established at the center of the chessboard, the horizontal right direction is the positive direction of x, and the vertical upward direction is the positive direction of y.

8. Contact net geometric parameters real-time measurement device based on single surveillance camera machine, its characterized in that: the pantograph control system comprises an image input module, a pantograph positioning module, a pantograph slide bar outline extraction module, a contact line extraction module, an intersection point acquisition module and a height guide and pull value calculation module; the image input module is used for receiving the image transmitted by the single monitoring camera and transmitting the received image to the pantograph positioning module; the pantograph positioning module is used for positioning a pantograph area image in the image received by the image input module and transmitting the reserved pantograph area image to the pantograph slide bar outline extraction module and the contact line extraction module;

9. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, which when executed in a computer processor, implements the steps of the method for measuring geometrical parameters of a catenary in real time based on a single monitoring camera according to any one of claims 1 to 7.

10. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for measuring geometrical parameters of a catenary based on a single monitoring camera in real time according to any one of claims 1 to 7.