CN110930415B - Method for detecting spatial position of track contact net - Google Patents

Abstract

The application provides a method for detecting the spatial position of a track contact net, which is applied to a photoelectric measurement system device 1 and comprises the following steps: acquiring an original image through an area-array camera 11 and a line laser 13, and sending the original image to an image computer 12; the image computer 12 obtains a connected domain where the contact network cable is located by performing image processing on the original image; segmenting the connected domain where the contact network cable is located by using an image segmentation method to obtain the region where the actual contact network cable is located; calculating the gray scale gravity center coordinate of the area where the actual contact network cable is located by using a gray scale gravity center method; converting the gray gravity center coordinates through image coordinates to obtain contact net space coordinates; and calculating the space position parameters of the track overhead line system according to the space coordinates of the overhead line system.

Description

A method for detecting the spatial position of a rail catenary

technical field

The invention belongs to the technical field of photoelectric detection, and in particular relates to a method for detecting the spatial position of a non-contact track catenary.

Background technique

The pantograph of the electric locomotive relies on the catenary erected along the railway to continuously obtain the electric energy for operation. Since the electric locomotive pantograph is in sliding contact with the contact wire, there will be a certain lifting force on the contact wire during normal operation, which will lead to separation from the contact wire when it reaches a certain speed, and the contact suspension has different degrees along the span. difference. The catenary is an open-air device, and it will shift or dance under the action of strong wind, and in some cases, it will cause a bow failure. If the pull-out value is too small, the purpose of uniform wear of the slide plate and prolonging the life of the pantograph cannot be achieved; if the pull-out value is too large, it is easy to cause bow scraping or drilling accidents. Therefore, in order to ensure the good contact between the pantograph and the catenary wire and the reliable current receiving, it is necessary to inspect the conductance value and pull-out value of the catenary.

For the special catenary detection train, the current non-contact detection method of the contact line guide height and pull-out value is the line array camera image processing method. The basic method is to use the binocular triangle method composed of two cameras and cooperate with the illumination of the supplementary light source. The working status images of the pantograph and contact wire are obtained by binocular recognition technology, and a certain image processing algorithm is used to process the images of the pantograph and contact wire, and the pull-out value and conductance of the contact wire are obtained. Although the above linear array camera detection method has high measurement accuracy, the environmental adaptability of the measurement device is poor, and the data analysis volume is large and the measurement speed is low.

Contents of the invention

The invention provides a photoelectric measurement system of a non-contact rapid subway catenary space position detection device, which realizes the detection of catenary guide height and pull-out value, and provides detection data.

The present application provides a method for detecting the spatial position of the rail catenary, the detection method is applied to a photoelectric measurement system device (1), and the photoelectric measurement system device (1) is set on the centerline position of the rail walking detection platform relative to the rail (1) The photoelectric measurement system device (1) comprises an area array camera (11), an image computer (12) and a line laser (13), and the area array camera (11) and the line laser (13) are connected with the image computer (12); The area array camera (11) is used to photograph the image of the rail catenary, and the line laser (13) is used for the illumination compensation light source of the photoelectric measurement system, and the detection method includes:

Gather original images by an area array camera (11) and a line laser (13), and send the original images to an image computer (12);

The image computer (12) obtains the connected domain where the catenary line is located by performing image processing on the original image;

Segmenting the connected domain where the catenary line is located by using an image segmentation method to obtain the area where the catenary line is actually located;

Calculate the gray-scale center-of-gravity coordinates of the area where the actual catenary line is located with the gray-scale center-of-gravity method;

Converting the coordinates of the gray center of gravity through image coordinates to obtain the spatial coordinates of the catenary;

The spatial position parameters of the catenary are calculated according to the spatial coordinates of the catenary.

Preferably, the image computer (12) obtains the connected domain where the catenary line is located by performing image processing on the original image, specifically including:

The image computer (12) performs median filter processing, global binarization processing and morphological processing on the original image to obtain the connected domain where the catenary line is located.

Preferably, the image segmentation method is used to segment the connected domain where the catenary line is located to obtain the actual area where the catenary line is located, specifically including:

The connected domain where the catenary line is located is segmented by using the gama transformation and the binarization method of the Otsu method to obtain the area where the catenary line is actually located.

Preferably, the image segmentation method is used to segment the connected domain where the catenary line is located, specifically including:

The image segmentation method is used to segment the connected domain where the catenary line is located by separating the catenary line and the image area of the catenary line installation base.

Preferably, the method also includes:

According to the optical structure parameters of the photoelectric measurement system device (1), the distance between the area array camera (11) and the line laser (13), the camera elevation angle and the lens focal length of the area array camera (11) are determined.

Preferably, said transforming said gray scale center of gravity coordinates through image coordinates to obtain catenary space coordinates specifically includes:

Calibrate the photoelectric measurement system device (1), and calculate the coordinate conversion calibration parameters of the gray scale center of gravity coordinates and catenary space coordinates;

The coordinates of the gray center of gravity are transformed through image coordinates according to the coordinate conversion calibration parameters to obtain the catenary space coordinates.

Preferably, said calibration of the photoelectric measurement system device (1) specifically includes:

The photoelectric measurement system device (1) is calibrated by using a two-axis linear slide table.

Preferably, the spatial position parameters of the track catenary include the pull-out value and height guide value of the track catenary.

Compared with the prior art, the technical features of the present invention that are different from the prior art solutions are:

1. Using image acquisition devices such as area array cameras and line lasers, using the two-axis linear sliding table calibration method, using the image processing algorithm to segment the actual catenary image area, the camera image is obtained through two-axis linear sliding table calibration and fitting. The mapping relationship between the two-dimensional coordinates of the catenary line and the spatial coordinates of the catenary line is used to obtain the spatial position of the catenary line.

2. The acquisition computer is miniaturized, and the power supply control is realized internally, and the cables are routed inside the detection device.

The invention is a catenary spatial position detection module in an intelligent rail detection vehicle, based on the optical structure method, a photoelectric measurement system composed of an area array camera, a line laser, and an image computer, and after accurate calibration, an image suitable for the structure is used The processing algorithm can realize the non-contact measurement of the spatial position of the catenary.

Description of drawings

Fig. 1 is the structure of a kind of catenary detection photoelectric measurement system provided by the embodiment of the present application;

Fig. 2 is a schematic diagram of calculating the installation position of a line laser and a camera provided by the embodiment of the present application;

Fig. 3 is a block diagram of the operation flow of a photoelectric measurement system provided by the embodiment of the present application;

Fig. 4 is a kind of data processing process based on total station calibration method provided by the embodiment of the present application;

FIG. 5 is an effect diagram of an original image of a catenary provided in an embodiment of the present application;

FIG. 6 is an effect diagram of catenary image processing provided by the embodiment of the present application;

Fig. 7 is an effect diagram of a division of catenary line connected areas provided by the embodiment of the present application;

In the figure: 1-photoelectric measurement system, 11-area array camera, 12-image computer, 13-line laser, 2-installation base, L-the horizontal distance between the front center of the camera lens and the plane of the laser, θ-the distance between the optical axis of the camera and the plane of the laser The angle between the horizontal plane, d0-camera working depth, d1-the distance from the front of the lens to the first measurement surface, d2-the distance from the front of the lens to the second measurement surface, h0-the height of the lower bound of the catenary guide height, h1-the upper bound of the catenary guide height Height, 0-camera lens front center point, 01-first reference point, 02-second reference point.

Detailed ways

In terms of hardware structure, the catenary detection device of the present invention is a photoelectric measurement system for catenary detection, which mainly includes: an area array camera, a line laser, and an image computer. As a contact line lighting device, the line laser solves the interference effect of natural light and improves the stability of the photoelectric measurement system. The area array camera shoots the catenary illuminated by the line laser to obtain the catenary image, which is an image acquisition device for catenary detection. As the control center and data storage center of the entire detection device, the image computer is used to control the acquisition of the camera and store the image data of the camera. The image computer is connected to the area array camera to collect the camera image and process it.

In relation to the installation structure of the photoelectric detection system, the installation mechanical structure includes camera and light source installation platform components and positioning components. Among them, the camera and light source installation platform components are in an inverted M shape, and the camera and line lasers are respectively installed on the support arms at both ends. Since the catenary is erected along the direction of the railway, the camera and the line laser are relatively installed on the installation platform assembly along the direction of the catenary, and the installation platform assembly is accurately positioned and installed on the detection platform of the inspection vehicle through the positioning component.

In the early stage of detection, the optical structure parameter design and system calibration of the photoelectric detection system of the present invention are carried out. Firstly, it is ensured that the area array camera can capture the guide height in the range of 4000-5300mm and the pull-out value within ±200mm in the above-mentioned installation position relationship. Catenary images within the scope and to ensure sufficient measurement accuracy. From this point of view, the camera pixel size, target surface size, etc. are taken as initial conditions for design. After determining the installation distance between the camera and the line laser, determine the installation angle range between the camera and the horizontal plane based on the measured guide height and the pull-out value range, and traverse within this angle range to determine the best spatial resolution of the system and the corresponding The focal length of the camera is used to determine the installation angle of the camera, and the optical structure parameters of the system can be obtained under the condition of ensuring the depth of field of the camera. Guarantee feasibility and detection accuracy.

In order to obtain the mapping relationship between the gray center coordinates of the camera image and the catenary space coordinates, this relationship can be used in the image processing in the later stage of the detection, so that the catenary conductance and pull-out value can be calculated directly from the camera image by using the formula. The system calibration method of the detection device is as follows: this system uses the two-axis linear sliding table calibration method. Within the measurement range of the equipment, the markers move on the sliding table at equal intervals along the direction of pulling out and height guide, and the markers move on the sliding table at equal intervals. The coordinate point set under the table coordinate system, and then use the total station to measure and calculate the conversion relationship between the equipment coordinate system of the sliding table and the catenary equipment coordinate system, and use the mapping calibration relationship to calculate the guide height and pull-out value of the sign in batches , to achieve rapid calibration.

The detection workflow of the photoelectric detection system for the catenary height and pull-out value is as follows: when the detection vehicle enters the running detection state, the area array camera, the image computer and the line laser are installed in the same relative position as the catenary line and put into work, and the line laser Illuminate the catenary line vertically above, and the area array camera shoots the illuminated section of the catenary line at a certain angle, and the grayscale image obtained by shooting is sent back to the image computer. After value filtering, global binarization, connected domain processing and segmentation, separation of the catenary line and the image area of the catenary line installation base, finally obtain the area where the actual catenary line is located, and calculate the gray center of the actual catenary area in the image coordinate system Coordinates, and then use the mapping formula obtained from the above calibration to calculate the spatial coordinates of catenary conduction height and pull-out value.

The present invention is a catenary spatial position detection module in a miniaturized dynamic intelligent track detection vehicle, and the technical scheme of the photoelectric measurement system of the catenary detection device involved is as follows:

First of all, the overall structure of the photoelectric measurement system of the catenary detection device is as follows:

With reference to Fig. 1, it is a kind of structure of the photoelectric measurement subsystem that the inventor designed to detect the catenary spatial position in the creation process of the present invention. A light source installation platform component and a positioning component. Among them, the camera and light source installation platform components are in an inverted M shape, and the camera and line lasers are respectively installed on the support arms at both ends. Since the catenary is erected along the direction of the railway, the camera and the line laser are relatively installed on the installation platform assembly along the direction of the catenary, and the installation platform assembly is accurately positioned and installed on the detection platform of the inspection vehicle through the positioning component. The mounting platform assembly provides a mounting base for various items in the photoelectric measurement device. The area array camera and the line laser are respectively installed in the mounting holes of the device; the image computer is placed in the base at the middle part, and the image computer used for acquisition is miniaturized.

Among them, the area array camera is controlled by the internal power supply of the detection device; the laser generator is placed on the side of the line laser installation arm, and the power supply control is also realized internally.

The area array camera, as an image acquisition device of the photoelectric measurement system, captures catenary images.

The image computer, as the control center and data storage center of the entire detection device, is used to control the acquisition of the camera and store the returned image data. The image computer is used to receive the image collected by the area array camera, and process and solve the image.

The installation platform assembly is arranged on the structure of the catenary detection photoelectric measurement system through the positioning parts, and the catenary detection photoelectric measurement system structure walks along the track on the detection platform, and through the area array camera, image computer, line laser and internal The cable detects the catenary along the direction of the track.

The function of the positioning component is to accurately install the photoelectric measurement system of the invention on the detection platform of the detection vehicle.

The overall installation relationship between the device and the inspection vehicle is as follows: the catenary is erected along the direction of the railway. In order to enable the camera and the line laser to take a complete picture of the moving catenary, the camera and the line laser should be installed on the installation platform opposite to each other along the direction of the catenary. On the assembly, the installation platform assembly is accurately positioned and installed on the inspection platform of the inspection vehicle through the positioning component. That is, the field of view of the camera and the laser is on the same plane, and the connection line is perpendicular to the line direction of the catenary. Because the line laser is installed vertically to illuminate the catenary line above the track, in order for the camera to capture the catenary line illuminated by the laser, the camera is installed at a certain angle to the horizontal plane and at a certain horizontal distance from the line laser.

Therefore, in order to ensure that the area array camera can capture the catenary image in the above-mentioned installation position relationship, and to ensure sufficient measurement accuracy from the internal optical parameters of the camera, it is necessary to determine the horizontal distance between the front center of the camera lens and the laser plane at the design stage That is, the distance between the two mounting holes of the device, the angle between the optical axis of the camera and the horizontal plane, and the focal length of the camera lens. The following is the installation and layout calculation process in the optical structure design stage:

Please refer to Figure 2, which is a schematic diagram of the calculation principle of the installation position of the line laser and the camera of the photoelectric measurement system of the present invention:

The principle of optical structure design and layout calculation is: design with the camera pixel size, target surface size, etc. as the initial conditions. After determining the installation distance between the camera and the line laser, determine the installation angle range between the camera and the horizontal plane based on the measured guide height and the pull-out value range, and traverse within this angle range to determine the best spatial resolution of the system and the corresponding The focal length of the camera is used to determine the installation angle of the camera, and the optical structure parameters of the system can be obtained under the condition of ensuring the depth of field of the camera. Guarantee feasibility and detection accuracy. The specific calculation process is as follows: when the inspection vehicle enters the running detection state, an area array camera, an image computer and a line laser are installed in the same relative position as the catenary line, and put into operation. The line laser illuminates the catenary line vertically above, and the area array camera compares The bright contact line is photographed, and the obtained camera image is sent back to the image computer, and the image computer uses the image processing algorithm, that is, binarization, filtering, connected domain segmentation processing of the camera image, separation of the catenary line and the image area of the catenary line installation base , and finally get the separation of the area where the actual catenary line is located, and get the gray center coordinates of the actual catenary area, and then use the mapping formula obtained by calibration to calculate the spatial coordinates of the catenary conductance height and pull-out value.

The user uses the above system to collect the work process as follows:

Referring to Fig. 3 and Fig. 4, it is a block diagram of the operation flow chart of the photoelectric measurement system and a flow chart of data processing. During the collection process, the working process of the above catenary space position detection device is as follows:

During the operation of the inspection vehicle, the laser illuminates the catenary above the laser, and the camera photographs the catenary above the line laser under the control of the image computer, and transmits the image to the image computer, and the image computer extracts the catenary feature points from the camera image And process, and then store the measurement results on the hard disk in a certain data storage format, and send the processing results to the user's comprehensive computer, generate output waveforms and reports, and provide historical data analysis and over-limit warning. The above realizes the collection and processing of catenary image data and the storage and transmission of the processed data.

The catenary detection system of the present invention is based on the gray center coordinates of the catenary image of the CCD camera, and is mapped to the spatial coordinates of the pull-out value of the catenary wire and the guide height value. Then the pull-out value and guide height value of catenary wire can be calculated.

The mathematical model on which catenary spatial position measurement relies is the following system calibration formula:

Height＝(p ₀ +p ₁ x+p ₂ x ² +p ₃ x ³ )(q ₀ +q ₁ y+q ₂ y ² +q ₃ y ³ )

Taggcr-d|p ₀ x|p ₁ x ² |p ₂ x ³ |q ₀ y|q ₁ y ² |q ₁ y ² |c ₀ xy|c ₁ x ² y ² |c ₂ x ² y|c _3xy2 ^_

In the formula, x and y are the coordinates of the tangent point under the catenary line in the image coordinate system, and p _i , q _i , c _i and d are the calibrated parameters.

During the above calibration process and the computer image processing process of the collected images, due to the presence of interference objects such as salt noise, gaps and holes, and catenary cable installation bases in the original camera image, it is necessary to perform image processing on the camera image to separate the actual catenary characteristic area. The algorithm for extracting catenary feature points from camera images by the image computer in the embodiment of the present invention will be further described in detail below.

Since the catenary image feature area is located at the bottom of the camera image, the following algorithm needs to be designed to find the lower edge position of the catenary line:

1. The image first uses median filtering to remove salt noise, then performs global binarization, uses closed operations to remove small gaps and holes, and finds connected domains;

2. Calculate the central coordinates of each connected domain, and sort the connected areas from large to small according to the y-axis coordinate value of the central point of the connected domain;

3. Using variables such as connected domain area and moment of inertia as screening conditions, traverse the sorted connected domain set in step 2, and remove the connected domains of interfering objects. The connected domain that can meet the filtering conditions and has the smallest subscript number is is the connected domain where the catenary line is located;

4. Segment the connected domain where the catenary line is located, and separate the catenary line and the image area of the catenary line installation base.

Referring to FIG. 5, FIG. 6, and FIG. 7, it shows the segmentation effect of the catenary line connected domain. The separation idea adopted is:

The middle and both ends of the connected domain image where the catenary line is located are bright, the middle part corresponds to the target catenary line, the left and right sides of the catenary line area, and the folded wing-shaped area correspond to the installation base of the catenary line. Since the lower end of the cross-section of the catenary line is a circle, the illuminance of the line laser on both sides of the catenary line is low, so there is a darker transition zone between the middle area where the catenary line is located and the areas on both sides in the image, as shown in It is shown in the darker part of Figure 5.

Perform a non-linear operation on the Gama transformation of Figure 5, so that the gray value of the input image and the output image have an exponential relationship. If the value of gama is greater than 1, the brighter area will be stretched, and the darker area will be compressed to become darker and enhanced. For the contrast of the image, the Otsu method is used to obtain the threshold and then binarized to obtain the image shown in Figure 6. At this time, the middle area and the two sides are interrupted. Figure 6 is further processed to find the left and right split columns of mounting bases and catenary lines on the image. Due to the poor symmetry of the light intensity distribution in the middle part, resulting in poor shape stability in the middle part, it is necessary to scan column by column starting from both sides of the image to find two split columns. The process of finding the split column is to count the number of pixels with a gray level of 1 in each column during the column-by-column scanning process. For the areas on both sides, the distribution of the number of 1-value pixels has two changes: low, high, and low. In this case, by identifying this pattern, the columns of the regions on both sides can be eliminated, and the columns of the catenary line image in the middle of the image can be obtained. The connected domain finally obtained according to this method is shown in the gray area in Figure 7.

5. Calculate the coordinates of the gray center of gravity of the area where the catenary line extracted above is located, and the calculation formula is:

In the formula, m and n are the number of rows and columns of the segmented detection line area respectively, x _i represents the i-th row coordinate, y _j represents the j-th column coordinate, f _ij represents the pixel gray value of the i-th row and j-th column .

Based on the above image processing algorithm, the area where the catenary line is located and its gray-scale center of gravity coordinates are obtained. During the calibration process, at the same time, the total station is used to measure the guide height and pull-out value of the mark point of the catenary mark, and the following can be used The formula fits each calibration parameter in the mapping formula;

Similarly, when the system collects image processing data, the system can automatically calculate the pull-out value and guide height value of the catenary line corresponding to the collection point by combining the image coordinates of the catenary after image processing with the completed calibration formula.

The mapping formula between image coordinates and space coordinates used in the software is:

Height＝(p ₀ +p ₁ x+p ₂ x ² +p ₃ x ³ )(q ₀ +q ₁ y+q ₂ y ² +q ₃ y ³ )

Tagger＝d+p ₀ x+p ₁ x ² +p ₂ x ³ +q ₀ y+q1y ² +q ₂ y ³ +c ₀ xy+c ₁ x ² y ² +c ₂ x ² y+c ₃ xy ²

In the formula, x and y are the coordinates of the tangent point under the catenary line in the image coordinate system, and p _i , q _i , c _i and d are constants. Its value is obtained after calibration.

After the above-mentioned image processing algorithm extracts the catenary feature points, the detection software will store the processed catenary image data and the calculated catenary spatial position data on the local hard disk of the image computer. At the same time, these data will be sent to the user's integrated computer, and the integrated computer will be aggregated and fused. The catenary detection device can output waveforms and reports online for catenary geometric parameters such as catenary height and pull-out value, and provide historical data analysis and over-limit early warning. It can not only detect hidden dangers in time, but also provide data for subway catenary maintenance , Effectively guarantee the maintenance and repair of railway lines.

Claims (8)

1. a track catenary space position detection method, it is characterized in that, described detection method is applied to photoelectric measurement system device (1), and photoelectric measurement system device (1) is arranged on track walking detection platform relative to track (1) On the center line, it is connected with the track walking detection platform through the installation base (2); the photoelectric measurement system device (1) includes an area array camera (11), an image computer (12) and a line laser (13), and the area array camera ( 11) and the line laser (13) are connected with the image computer (12); the area array camera (11) is used to take the image of the rail catenary, and the line laser (13) is used for the illumination compensation light source of the photoelectric measurement system, so The detection methods include: Gather original images by an area array camera (11) and a line laser (13), and send the original images to an image computer (12); The image computer (12) obtains the connected domain where the catenary line is located by performing image processing on the original image; Segmenting the connected domain where the catenary line is located by using an image segmentation method to obtain the area where the catenary line is actually located; Calculate the gray-scale center-of-gravity coordinates of the area where the actual catenary line is located with the gray-scale center-of-gravity method; Converting the coordinates of the gray center of gravity through image coordinates to obtain the spatial coordinates of the catenary; The spatial position parameters of the catenary are calculated according to the spatial coordinates of the catenary. 2. detection method according to claim 1, is characterized in that, described image computer (12) obtains the connected domain where catenary line is located by carrying out image processing to original image, specifically comprises: The image computer (12) performs median filter processing, global binarization processing and morphological processing on the original image to obtain the connected domain where the catenary line is located. 3. The detection method according to claim 1, wherein the connected domain where the catenary line is located is segmented using an image segmentation method to obtain the area where the catenary line is actually located, specifically comprising: The connected domain where the catenary line is located is segmented by using the gama transformation and the binarization method of the Otsu method to obtain the area where the catenary line is actually located. 4. The detection method according to claim 1, wherein the segmenting the connected domain where the catenary line is located by using the image segmentation method specifically comprises: The image segmentation method is used to segment the connected domain where the catenary line is located by separating the catenary line and the image area of the catenary line installation base. 5. detection method according to claim 1, is characterized in that, described method also comprises: According to the optical structure parameters of the photoelectric measurement system device (1), the distance between the area array camera (11) and the line laser (13), the camera elevation angle and the lens focal length of the area array camera (11) are determined. 6. The detection method according to claim 1, wherein said converting the gray-scale center-of-gravity coordinates to obtain the catenary space coordinates through image coordinate conversion specifically includes: Calibrate the photoelectric measurement system device (1), and calculate the coordinate conversion calibration parameters of the gray scale center of gravity coordinates and catenary space coordinates; The coordinates of the gray center of gravity are transformed through image coordinates according to the coordinate conversion calibration parameters to obtain the catenary space coordinates. 7. The detection method according to claim 6, wherein said calibration of the photoelectric measurement system device (1) specifically includes: The photoelectric measurement system device (1) is calibrated by using a two-axis linear slide table. 8. The detection method according to claim 1, characterized in that, the spatial position parameters of the track catenary include the pull-out value and the height guide value of the track catenary.

Images