Device and method for imaging the connection area between the contact network elastic sling and the load-bearing cable

Abstract

The present invention discloses an imaging device and imaging method for the connection area between the elastic sling of the contact network and the supporting cable. The device comprises: a trigger unit and an imaging unit placed on the roof of a vehicle, and an electrical control unit and a detection host placed in the vehicle; the detection host is respectively connected to the trigger unit and the imaging unit; the trigger unit is connected to the imaging unit; the electrical control unit is respectively connected to the detection host, the trigger unit and the imaging unit; the method comprises: identification of the connection point between the contact network elastic sling and the supporting cable, and imaging of the connection area between the contact network elastic sling and the supporting cable. The present invention solves the problem that the traditional inspection equipment for the suspension state of the contact network does not have the capability to image the connection area between the elastic sling of the contact network and the supporting cable, and can collect high-definition imaging of the connection area between the elastic sling of the contact network and the supporting cable, and can distinguish whether there is a break or detachment fault at the connection point between the elastic sling and the supporting cable, thus providing a guiding basis for the maintenance of the suspension state of the contact network.

Description

Technical Field

[0001] The invention relates to the technical field of electrified railway contact network detection, and in particular to an imaging device and an imaging method for the connection area between an elastic sling and a load-bearing cable of the contact network. Background Art

[0002] With the rapid development of urban rail transit in my country, the overhead contact network provides power supply for trains of electrified urban rail transit. As the power source of electric locomotives, the safety of the overhead contact network is one of the important guarantees for the safety of high-speed train operations. The overhead contact network is a power supply network laid along the railway, which transmits electric energy through sliding contact between the contact line and the pantograph above the high-speed EMU. The static position of the contact line in space affects the dynamic current-collecting performance of the pantograph and the overhead contact network, as well as the vibration load of the contact network components. Standard TB10758-2018 "Construction Quality Acceptance Standard for High-speed Railway Electric Traction Power Supply Project" has strict regulations on the static spatial position of the contact line, such as: the allowable deviation of the contact line pull-out value is ±30mm, the allowable deviation of the height of the contact line suspension point from the rail surface is ±30mm, and the height difference of the contact line at two adjacent suspension strings within one span shall not be greater than 10mm. In the design stage of the overhead contact network, it is necessary to accurately calculate the geometric shape, line length, suspension string length, and positioner slope of the load-bearing cable and elastic suspension cable according to the design value of the static spatial position of the contact line, and prefabricate the suspension string, elastic suspension cable, support and positioning structure.

[0003] The traditional contact network status inspection device does not have the function of imaging the equipment in the connection area between the elastic sling (referred to as "elastic sling") and the load-bearing cable near the positioning point. However, as the operation and maintenance progressed, it was found that the wire clamps in the connection area between the elastic sling and the load-bearing cable of the contact network were broken, which seriously affected the safety of the contact network and threatened the operation of the train.

[0004] Patent application number CN201910722075.3 discloses a three-dimensional rapid shape-finding method for contact network suspension based on the static spatial position of the contact line. This method takes into account the actual stress state and elastic deformation of the contact network suspension line. It is simple and efficient, and can accurately calculate the three-dimensional static shape of the simple chain shape and elastic chain-shaped suspension of various line conditions and the actual static spatial position of the contact line. It is directly applied to the design of the bow-net system, the pre-configuration of the contact network construction, and the maintenance of the contact network status. However, this solution can only form a large amount of calculation data by calculating the three-dimensional rapid shape-finding of the contact network suspension, but cannot form a three-dimensional image. In the subsequent maintenance and repair process, personnel with extremely strong professional knowledge are required to understand the data, and it cannot provide an intuitive basis for maintenance and repair.

[0005] Patent application number CN201910831420.7 discloses a device and method for detecting the dynamic form of the catenary suspension string, wherein the device includes: an imaging module for detecting the dynamic form of the suspension string at the contact point between the contact line and the pantograph, and obtaining image data of the dynamic form of the suspension string; a light source module for imaging and supplementing the position of the suspension string; the imaging module and the light source module are arranged in the pantograph base area on the top of the vehicle; an equipment control module for controlling the synchronization of the shutter trigger of the imaging module and the light source opening and closing of the light source module; a data processing module for identifying the state of the suspension string according to the image data of the dynamic form of the suspension string, and obtaining the detection result of the dynamic form of the suspension string. The device and method can realize high-definition imaging of the dynamic form of the suspension string, and accurately identify the state of the suspension string. The acquisition and identification of the whole process can be completed automatically, with high processing efficiency, and the form of the suspension string under dynamic load can be fully grasped, which can provide effective information support for the prediction of the state of the suspension string. However, this method does not propose an effective solution for the imaging identification of the catenary elastic suspension and the load-bearing cable area.

[0006] For this purpose, it is necessary to develop an imaging detection method (device) for the contact network elastic sling and load-bearing cable area to monitor the status of the connection area between the contact network elastic sling and the load-bearing cable, detect faults in time and ensure the safety of the contact network. Summary of the invention

[0007] The purpose of the present invention is to overcome the shortcomings of the prior art and to provide an imaging device and an imaging method for the connection area between the elastic sling and the load-bearing cable of a contact network.

[0008] The objective of the present invention is achieved through the following technical solutions:

[0009] A device for imaging the connection area between an elastic suspension cable of a contact network and a load-bearing cable comprises a trigger unit and an imaging unit placed on a vehicle roof; and an electrical control unit and a detection host placed in the vehicle; the detection host is respectively connected to the trigger unit and the imaging unit; the trigger unit is connected to the imaging unit; the electrical control unit is respectively connected to the detection host, the trigger unit and the imaging unit for power supply.

[0010] Furthermore, the trigger unit includes a target detection and identification component and a trigger module; the target detection and identification component and the trigger module are both connected to the detection host, the trigger module is connected to the imaging unit, the target detection and identification component takes real-time and high-speed images of the connection area between the contact network elastic sling and the load-bearing cable and transmits them to the detection host, and when the detection host recognizes the target, the trigger module is started.

[0011] Furthermore, the target detection and identification component is a light section detection component based on machine vision technology, including a line light source device and a high-speed high-definition camera; the line light source device and the high-speed high-definition camera are installed on the roof through a fixing device; the line light source device is installed at an angle upward so that the light source is projected upward onto the load-bearing cable; the high-speed high-definition camera is connected to the in-vehicle detection host through a network, is installed at an angle upward, captures images of the elastic slings and load-bearing cables illuminated by the line light source, and transmits them to the detection host.

[0012] Furthermore, the fixing device is installed on the edge of the roof in a parallel line direction without exceeding the vehicle limit; and is fixed to the roof through three mounting seats.

[0013] Furthermore, the imaging unit includes two groups of imaging components installed on the roof in a centrally symmetrical manner, and the imaging components include a high-definition imaging camera and a xenon lamp.

[0014] Furthermore, the method comprises the following steps:

[0015] Identification of the connection points between the overhead wire and the catenary:

[0016] Use a high-definition high-speed camera to collect bitmap images of the connection points between the overhead line crane and the catenary cable;

[0017] Preprocess the image;

[0018] Identify the image blob and calculate the blob area. Preliminarily select the target image based on the feature that the blob area of ​​the connection area between the suspension string and the catenary cable or the image blob of only the catenary cable is significantly smaller than the blob area of ​​the elastic suspension cable and the catenary cable area.

[0019] Perform feature matching based on multiple templates to identify the connection point between the overhead line crane and the catenary cable;

[0020] Imaging of the connection area between the overhead line spring and the catenary:

[0021] After the target recognition software of the contact network hanging and the connecting point of the catenary cable recognizes the target, it immediately controls the trigger module to send a falling edge trigger pulse;

[0022] The high-definition imaging camera and xenon lamp enter the acquisition state instantly after receiving the falling edge trigger pulse, and acquire high-definition images of the target area;

[0023] The high-definition images are transmitted via the gigabit network to the in-vehicle detection host for processing, compressed into JPG images, and integrated with location information such as kilometer markers and station areas and stored on the hard disk.

[0024] Furthermore, the preprocessing includes performing a Y-direction difference on the image, binarizing the image according to a suitable threshold, dilating the binarized image, and connecting the white areas of the binarized image.

[0025] The beneficial effects of the present invention are as follows: it solves the problem that the traditional contact network status inspection equipment is not capable of imaging the connection area between the contact network elastic slings and the load-bearing cables. It can collect high-definition images of the connection area between the contact network elastic slings and the load-bearing cables, and can distinguish whether there are breakage or detachment faults at the connection points between the elastic slings and the load-bearing cables, thus providing guidance for contact network status maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a block diagram of the device principle of the present invention.

[0027] Figure 2 It is a schematic diagram of the installation of the roof equipment of the present invention.

[0028] Explanation of reference numerals: 1 - high-speed high-definition camera; 2 - line light source equipment; 3 - high-definition imaging camera; 4 - xenon lamp. DETAILED DESCRIPTION

[0029] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the fact that ordinary technicians in the field can implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0032] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0033] In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the fact that ordinary technicians in the field can implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0034] Embodiment 1, as Figure 1 and Figure 2As shown, a device for imaging the connection area between an elastic suspension cable of a contact network and a load-bearing cable comprises a trigger unit and an imaging unit placed on the roof of a vehicle; and an electrical control unit and a detection host placed in the vehicle; the detection host is respectively connected to the trigger unit and the imaging unit; the trigger unit and the imaging unit are connected; the electrical control unit is respectively connected to the detection host, the trigger unit and the imaging unit for power supply.

[0035] Among them, the trigger unit includes a target detection and identification component and a trigger module; the target detection and identification component and the trigger module are both connected to the detection host, the trigger module is connected to the imaging unit, the target detection and identification component takes real-time and high-speed images of the connection area between the contact network elastic sling and the load-bearing cable and transmits them to the detection host, and when the detection host recognizes the target, the trigger module is started.

[0036] Among them, the target detection and recognition component is a light section detection component based on machine vision technology, including a line light source device 2 and a high-speed high-definition camera 1; the line light source device 2 and the high-speed high-definition camera 1 are installed on the roof through a fixing device; the line light source device 2 is installed upwardly at an angle so that the light source is projected upward onto the load-bearing cable; the high-speed high-definition camera 1 is connected to the in-vehicle detection host through a network, installed upwardly at an angle, and captures images of the elastic slings and load-bearing cables illuminated by the line light source, and transmits them to the detection host.

[0037] Wherein, the fixing device is installed on the edge of the roof in parallel with the line direction without exceeding the vehicle limit; and is fixed to the roof through three mounting seats.

[0038] The imaging unit includes two sets of imaging components mounted on the roof in a centrally symmetrical manner to realize front and back imaging of the target area; the imaging components include a high-definition imaging camera 3 and a xenon lamp 4. When the elastic crane catenary cable connection point identification trigger unit identifies the elastic crane catenary cable connection point, the trigger module synchronously triggers the high-definition camera and the xenon lamp to perform high-definition imaging of the connection area between the elastic crane and the catenary cable.

[0039] Embodiment 2, wherein the installation positions of the high-speed high-definition camera 1, the linear light source device 2, the high-definition imaging camera 3, and the xenon lamp 4 are not limited to the positions of Embodiment 1.

[0040] A device for imaging the connection area between an elastic suspension cable and a load-bearing cable of a contact network comprises a trigger unit and an imaging unit arranged on a vehicle roof; and an electrical control unit and a detection host arranged in the vehicle.

[0041] Among them, the detection host is connected to the trigger unit and the imaging unit respectively; the trigger unit and the imaging unit are connected; the electrical control unit is connected to the detection host, the trigger unit and the imaging unit respectively for power supply.

[0042] Among them, the trigger unit includes a target detection and identification component and a trigger module; the target detection and identification component and the trigger module are both connected to the detection host, the trigger module is connected to the imaging unit, the target detection and identification component takes real-time and high-speed images of the connection area between the contact network elastic sling and the load-bearing cable and transmits them to the detection host, and when the detection host recognizes the target, the trigger module is started.

[0043] Among them, the target detection and recognition component is a light section detection component based on machine vision technology, including a line light source device 2 and a high-speed high-definition camera 1; the line light source device 2 and the high-speed high-definition camera 1 are installed on the roof through a fixing device.

[0044] The line light source device 2 is installed upwardly at an angle so that the light source is projected upwardly onto the load-bearing cable; the high-speed high-definition camera 1 is connected to the in-vehicle detection host through a network, installed upwardly at an angle, and captures images of the elastic slings and load-bearing cables illuminated by the line light source, and transmits them to the detection host.

[0045] Wherein, the fixing device is installed on the edge of the roof in parallel with the line direction without exceeding the vehicle limit; and is fixed to the roof through three mounting seats.

[0046] Among them, the fixing device can be installed on any side of the two edges. It only needs to ensure that the high-speed high-definition camera 1 and the linear light source device 2 can be projected directly on the position of the load-bearing cable. When the trolley runs along the railway, the cable line of the entire load-bearing cable can be identified.

[0047] The imaging unit includes two sets of imaging components mounted on the roof in a centrally symmetrical manner to realize front and back imaging of the target area; the imaging components include a high-definition imaging camera 3 and a xenon lamp 4. When the elastic crane catenary cable connection point identification trigger unit identifies the elastic crane catenary cable connection point, the trigger module synchronously triggers the high-definition camera and the xenon lamp to perform high-definition imaging of the connection area between the elastic crane and the catenary cable.

[0048] This device is suitable for overhead line operation vehicles, engineering vehicles, inspection vehicles and other engineering vehicles.

[0049] The imaging method of the connection area between the contact network elastic sling and the load-bearing cable comprises the following steps:

[0050] Identification of the connection points between the overhead wire and the catenary:

[0051] 1. Use a high-definition high-speed camera 1 to collect a bitmap image of the connection point between the overhead line spring and the load-bearing cable;

[0052] 2. Preprocess the image and filter out some interference signals.

[0053] 3. Identify image blobs;

[0054] 4. Calculate the blob area, and preliminarily select the target image by taking into account the fact that the blob area of ​​the connection area between the suspension string and the load-bearing cable or the image of only the load-bearing cable is significantly smaller than the blob area of ​​the elastic suspension cable and the load-bearing cable area;

[0055] 5. To further confirm that this image contains the connection between the elastic sling and the load-bearing cable, it is necessary to further confirm the target image selected above. At this time, feature matching is performed based on the multiple templates produced. When the matching degree reaches more than 90%, it can be confirmed that the image contains the target of "the connection point between the contact network elastic sling and the load-bearing cable";

[0056] Imaging of the connection area between the overhead line spring and the catenary:

[0057] After the target recognition software of the contact network hanging and the connecting point of the catenary cable recognizes the target, it immediately controls the trigger module to send a falling edge trigger pulse;

[0058] After receiving the falling edge trigger pulse, the high-definition imaging camera 3 and the xenon lamp 4 instantly enter the acquisition state to acquire high-definition images of the target area;

[0059] The high-definition images are transmitted to the in-vehicle detection host via a gigabit network for processing, compressed into JPG images, and integrated with location information such as kilometer markers and station areas and stored on the hard disk.

[0060] The preprocessing includes performing Y-direction difference on the image, binarizing the image according to a suitable threshold, dilating the binarized image, and connecting the white areas of the binarized image.

[0061] Blob in computer vision refers to a connected area in an image. Blob analysis is to perform connected domain analysis on a binary image after foreground / background separation.

[0062] Extraction and marking. Each marked Blob represents a foreground target, and then some relevant features of the Blob can be calculated. The core idea of ​​the Blob algorithm is to find the range of "grayscale mutation" in an area. Determine its size, shape and area, etc. The algorithm starts from the "convolution algorithm" in edge search. For example, the matrix of the "convolution algorithm" is set to 2*4 and scans from left to right and from top to bottom. Every time the matrix moves one step, the algorithm calculates the difference between the sum of the grayscale values ​​of the front half of the matrix (columns 1 and 2) and the sum of the grayscale values ​​of the back half of the matrix (columns 3 and 4). If the color of this image in the matrix is ​​the same, that is, the grayscale of the 8 pixels in the matrix is ​​very close, then the difference in the total grayscale value of the front and back of the matrix should be close to 0. The program runs step by step, and the difference in the total grayscale value of the front and back of the matrix is ​​always close to 0. If the matrix scans the edge of a foreground, the difference in the total grayscale value of the front and back of the matrix suddenly changes, and the grayscale value is no longer close, then a Blob has been reached. When the algorithm scans the entire image, the system records the coordinates of all the points where the mutation occurs. After that, a series of analyses are performed on these edge points to obtain information such as the size, shape, and area of ​​the blob.

[0063] This device uses a target recognition trigger unit to collect images of the contact network hanger and supporting cable area, and uses an edge matching image processing algorithm for identification. When the connection point between the hanger and the supporting cable is identified, the trigger module quickly triggers the high-definition imaging of each device in the imaging unit, and uses deep learning to automatically identify defects in the collected high-definition images.

[0064] The present invention can realize intelligent identification of defects at the connection points between the overhead line spring hanger and the catenary, and the process is as follows:

[0065] 1. Image acquisition;

[0066] 2. Image annotation and cleaning of training and test data sets;

[0067] 3. Neural network and loss function selection for detecting the connection points between the spring crane and the load-bearing cable;

[0068] The target detection network generally uses YOLO or SSD;

[0069] The loss function selects the SoftMax classification loss function;

[0070] 4. Target detection neural network training and effect testing;

[0071] 5. Defect recognition neural network and loss function selection;

[0072] The defect recognition network generally uses the GAN network;

[0073] 6. Training and testing of defect detection neural networks;

[0074] 7. Practical application, real-time identification of defects.

[0075] The present invention solves the problem that traditional contact network status inspection equipment is not capable of imaging the connection area between the contact network elastic sling and the load-bearing cable. The present invention can collect high-definition images of the connection area between the contact network elastic sling and the load-bearing cable, and can distinguish whether there is a breakage or detachment fault at the connection point between the elastic sling and the load-bearing cable, thereby providing guidance for contact network status maintenance.

[0076] The structure, function and connection form disclosed in the present invention can be realized in other ways. For example, the embodiments described above are only exemplary; in addition, each functional component in each embodiment of this invention can be integrated into one functional component, or each functional component can exist physically separately, or two or more functional components can be integrated into one functional component.

[0077] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements fall within the scope of the present invention to be protected. The scope of the present invention to be protected is defined by the attached claims and their equivalents.

Claims

1. An imaging device for the connection area between the contact network elastic sling and the load-bearing cable, characterized in that: It includes a trigger unit and an imaging unit placed on the roof of the vehicle, and an electrical control unit and a detection host placed in the vehicle; the detection host is connected to the trigger unit and the imaging unit respectively; the trigger unit is connected to the imaging unit; the electrical control unit is connected to the detection host, the trigger unit and the imaging unit respectively; The trigger unit includes a target detection and recognition component and a trigger module; the target detection and recognition component and the trigger module are both connected to the detection host, the trigger module is connected to the imaging unit, the target detection and recognition component takes a real-time high-speed image of the connection area between the contact network elastic sling and the load-bearing cable and transmits it to the detection host, and when the detection host recognizes the target, the trigger module is started; The target detection and recognition component is a light section detection component based on machine vision technology, including a line light source device and a high-speed high-definition camera; the line light source device and the high-speed high-definition camera are installed on the roof through a fixing device; the line light source device is installed upwardly tilted so that the light source is projected upward onto the catenary; The high-speed high-definition camera is connected to the in-vehicle detection host through a network, is installed obliquely upward, captures images of the elastic sling and the load-bearing cable illuminated by the linear light source, and transmits the images to the detection host; The imaging unit includes two groups of imaging components which are centrally symmetrically mounted on the roof, and the imaging components include a high-definition imaging camera and a xenon lamp.

2. The imaging device for the connection area between the contact network elastic sling and the load-bearing cable according to claim 1, characterized in that: The fixing device is installed on the edge of the roof in parallel with the line direction without exceeding the vehicle limit; and is fixed to the roof through three mounting seats.

3. The imaging method of the device for imaging the connection area between the contact network elastic sling and the catenary according to any one of claims 1-2, characterized in that: The method comprises the following steps: Identification of the connection points between the overhead wire and the catenary: Use a high-definition high-speed camera to collect bitmap images of the connection points between the overhead line crane and the catenary cable; Preprocessing the image; the preprocessing includes performing a Y-direction difference on the image, binarizing the image according to a suitable threshold, dilating the binarized image, and connecting the white area of ​​the binarized image; Identify the image blob and calculate the blob area. Preliminarily select the target image based on the feature that the blob area of ​​the connection area between the suspension string and the catenary cable or the image blob of only the catenary cable is significantly smaller than the blob area of ​​the elastic suspension cable and the catenary cable area. Perform feature matching based on multiple templates to identify the connection point between the overhead line crane and the catenary cable; Imaging of the connection area between the overhead line spring and the catenary: After the target recognition software of the contact network hanging and the connecting point of the catenary cable recognizes the target, it immediately controls the trigger module to send a falling edge trigger pulse; The high-definition imaging camera and xenon lamp enter the acquisition state instantly after receiving the falling edge trigger pulse, and acquire high-definition images of the target area; The high-definition images are transmitted via the gigabit network to the in-vehicle detection host for processing, compressed into JPG images, integrated with kilometer markers and station area location information, and then stored on the hard disk.

Images