CN107743577B

CN107743577B - Method and apparatus for detecting vibration information of electric railway vehicle

Info

Publication number: CN107743577B
Application number: CN201680035120.2A
Authority: CN
Inventors: 朴暎; 李基源; 赵容铉; 权三荣; 朴哲民
Original assignee: Korea Railroad Research Institute KRRI
Current assignee: Korea Railroad Research Institute KRRI
Priority date: 2015-06-15
Filing date: 2016-03-21
Publication date: 2020-06-09
Anticipated expiration: 2036-03-21
Also published as: CN107743577A; WO2016204385A1

Abstract

According to an embodiment of the present invention, a method for detecting vibration information of an electric railway vehicle based on a pantograph includes the steps of: determining a pantograph template; comparing the pantograph template with an input image obtained with a camera to detect a pantograph; and detecting vibration information of the pantograph based on a change in the position of the pantograph detected by comparison of each frame of the input image and the pantograph template.

Description

Method and apparatus for detecting vibration information of electric railway vehicle

Technical Field

The present invention relates to a method and apparatus for detecting vibration information of an electric railway vehicle based on a pantograph.

Background

Generally, a pantograph is a core device of an electric railway vehicle, which is in contact with a contact wire to supply electric power to the vehicle. The pantograph is mechanically contacted with a contact line for supplying electric energy during running, thereby generating abrasion and vibration, and generating an arc discharge phenomenon in which electric energy is discharged when the pantograph is not in contact with the contact line. In particular, the abnormal vibration is caused by various causes such as abnormality or overspeed of the pantograph, abnormality of the line, etc., causing the pantograph and the contact line not to contact, thereby generating an arc discharge phenomenon. In addition, the abnormal vibrations may fatigue the contact wires, thereby causing fatigue to components associated with the contact lines, resulting in damage to the contact lines and associated components.

The contact wire is a facility for stably supplying power to the railway vehicle body, and it is necessary to detect deformation and breakage of the contact wire caused by vehicles or external factors to maintain an optimum state according to the regulations of the electric railway facility.

The horizontal distance of the contact line from the center plane of the rail is called a pullout value, and if the pullout value of the contact line is too large, the current collector is detached from the contact line, causing an accident. Therefore, the pull-out value of the contact line is specified to have a certain limit, and generally the pull-out value of the contact line is required to be within 250mm from the center plane perpendicular to the rail. In general, even when the contact line is a straight line, the current collector needs to be dispersed on the effective surface to collect current in order to make the wear of the pantograph uniform, and therefore, a left and right pull-out value from the center of the rail is set for the facility. Such a contact wire is constituted by a contact wire which is in direct contact with an electric vehicle, and a support member, a power transmission line, a return line, and the like which support the contact wire. The contact line is a portion that supplies electric power to the electric vehicle, on which high-voltage current always flows, and therefore, its performance must be excellent. A dynamic pull-out value indicating an alignment state of a contact point between a train and a contact line on an electric railway is one of important criteria for determining safety performance of the train as one of criteria that can determine whether the train is stably supplied with current.

As a prior art document associated with this, korean patent laid-open No. 2014-0111712 (pantograph measurement method and pantograph measurement device) discloses a method of acquiring an image of a mark provided on a pantograph from a line sensor, generating a construction image from the acquired image, measuring the position of the pantograph by image processing to estimate the height position of a contact line.

However, the conventional pantograph measurement method based on image processing measures the position of the pantograph only with the laser sensor, and thus cannot obtain the dynamic pullout value of the contact line.

As a conventional method of sensing abnormal vibration, there is a method of detecting vibration of a pantograph by attaching a sensor for sensing vibration, such as an accelerometer, to the pantograph, insulating the sensor, and measuring the sensor. However, since the pantograph transmits electric power to the vehicle, it is always carried with a high voltage and a large current, and thus it is difficult to mount a sensor for vibration sensing.

Further, as a method of detecting the contact position of the pantograph and the contact wire, there is a method of mounting a video recorder on the upper portion of the vehicle so as to be able to detect the contact position of the pantograph and the contact wire for visual confirmation. However, since it is judged by naked eyes, it is difficult to derive an accurate quantitative value.

Regarding the method of monitoring the contact state of the pantograph and the contact wire, korean patent laid-open No. 1058179 (title of the invention: pantograph defect monitoring system), which is a prior art document, discloses a technique of comparing a current image of the pantograph, which is obtained by removing noise in an image obtained by a camera, with a reference image to determine whether or not there is a defect. Further, japanese patent No. 5534058 (title of the invention: wear measuring device and method thereof) as a prior art document discloses a wear measuring device and method thereof capable of obtaining a wear measurement value of a contact line in consideration of an inclination angle of a pantograph.

Disclosure of Invention

Technical problem

In order to solve the above-described problems, the present invention aims to provide a method of detecting the position of a pantograph and the position of a contact line without being affected by a luminance condition using a general camera capable of acquiring continuous images to improve the accuracy of measurement of a dynamic pullout value.

Another object of the present invention is to provide a method and an apparatus for detecting a pantograph by comparing a captured image with a pantograph template and detecting pantograph vibration of an electric railway vehicle by using an image processing technique.

However, the technical problems that the embodiments of the present invention are intended to achieve are not limited to the above technical problems, and other technical problems may also be present.

Technical scheme

In order to achieve the above object, a method for detecting vibration information of an electric railway vehicle based on a pantograph according to an embodiment of the present invention includes the steps of: determining a pantograph template; comparing the pantograph template with an input image obtained with a camera to detect a pantograph; and detecting vibration information of the pantograph based on a change in the position of the pantograph detected by comparison of each frame of the input image and the pantograph template.

Further, an apparatus for detecting vibration information of an electric railway vehicle based on a pantograph according to an embodiment of the present invention includes: a communication module that receives a captured image of a pantograph; a memory that stores a vibration detection program of the pantograph; and a processor that executes a vibration detection program of the pantograph, the processor executing the program to determine a pantograph template, compare the pantograph template with an input image obtained with the camera to detect the pantograph, and detect vibration information of the pantograph based on a change in a position of the pantograph detected by the comparison of each frame of the input image and the pantograph template.

Advantageous effects

According to any one of the above-described technical solutions of the present invention, there is an advantage of an additional illumination apparatus that accurately detects a dynamic pull-out value without uniformly adjusting the brightness of an image. Further, there is an advantage of utilizing a marker for narrowing the detection area of the pantograph to reduce the amount of calculation.

Further, in the case of an embodiment of the present invention, by detecting pantograph vibration using an image processing technique in an existing video monitoring method, additional equipment installation costs can be reduced, thus having very high economic feasibility.

Further, in the case of one embodiment of the present invention, since the pantograph is different in the abnormality of the slide plate, the abnormality of the track, the abnormality due to the change in the tension of the contact line, and the like depending on the natural frequency, the cause of the abnormality during the running of the electric railway vehicle can be determined, and the suspicious portion can be easily checked as necessary after the running.

Drawings

Fig. 1a is a block diagram of a contact line dynamic pull-out value detection method unaffected by brightness variation according to a first embodiment of the present invention.

Fig. 1b is a block diagram showing detailed steps constituting step S10 in fig. 1 a.

Fig. 2 is a contact line dynamic pull-out value detecting apparatus according to a first embodiment of the present invention.

Fig. 3 is a flowchart illustrating a process of calculating a feature vector and a step of classifying a reference image into different categories with reference to the feature vector according to a first embodiment of the present invention.

Fig. 4 is a schematic diagram of reference images classified into different categories according to the first embodiment of the present invention.

Fig. 5 is a schematic diagram of a process of updating and resetting divided categories according to the first embodiment of the present invention.

Fig. 6 shows a case where the obtained feature vectors differ by categories according to the first embodiment of the present invention.

Fig. 7 is a schematic diagram of the detection section detecting the pantograph from the pantograph template according to the first embodiment of the present invention.

Fig. 8 is a diagram for detecting the condition of a contact line by fusing hough transform and binarization according to the first embodiment of the present invention.

Fig. 9a shows a correctly detected pantograph image according to the first embodiment of the present invention.

Fig. 9b shows an incorrectly detected pantograph image according to the first embodiment of the present invention.

Fig. 10a shows a process of detecting a contact surface and a center point using an attached mark according to a first embodiment of the present invention.

Fig. 10b shows an attached label according to the first embodiment of the present invention.

Fig. 11 is a configuration diagram of a vibration detection device for a pantograph of an electric railway vehicle according to a second embodiment of the present invention.

Fig. 12 is a diagram for explaining a method of detecting a pantograph by each image frame according to the second embodiment of the present invention.

Fig. 13 is a diagram for explaining a method of measuring an angle between a horizontal line and a sliding plate in a detected pantograph according to a second embodiment of the present invention.

Fig. 14 is a diagram for explaining a method of calculating the vibration frequency of the pantograph based on an angle according to the second embodiment of the present invention.

Fig. 15a and 15b are diagrams for explaining a method of calculating the vibration intensity of the pantograph based on the angle magnitude according to the second embodiment of the present invention.

Fig. 16 is a view for explaining a method of measuring a horizontal line and left and right displacements of a slide plate according to the second embodiment of the present invention.

Fig. 17 is a diagram for explaining a method of calculating a frequency based on an average value of left and right displacements according to the second embodiment of the present invention.

Fig. 18 is a flowchart for explaining a method of detecting vibration of a pantograph of an electric railway vehicle according to a second embodiment of the present invention.

Detailed Description

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the embodiments. The invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Moreover, in order to clearly explain the present invention, portions that are not relevant to the description are omitted in the drawings, and like reference numerals are given to like portions throughout the specification.

Throughout the specification, when a certain portion is "connected" to another portion, not only a case of "directly connecting" but also a case of "electrically connecting" with other elements interposed therebetween are included. Further, when a part "includes" a certain constituent element means that other constituent elements are not excluded but other constituent elements may be included without particular mention to the contrary, it is to be understood that the existence or addition possibility of one or more other features, numerals, steps, actions, constituent elements, components, or a combination thereof is not previously excluded.

The present invention relates to a method and apparatus for detecting vibration information of an electric railway vehicle based on a pantograph, which determines a pantograph template, detects the pantograph by comparing the pantograph template with an input image acquired by a camera, and detects vibration information of the pantograph based on a change in position of the pantograph detected by the comparison of each frame of the input image with the pantograph template.

The present invention is divided into two embodiments, and the first embodiment relates to a method of detecting a dynamic pullout value of a contact line, and more particularly, to a method of obtaining an image by a measuring camera and a sensor provided at the upper end of a railway, analyzing the obtained image, and detecting a dynamic pullout value between a pantograph and a contact line as information on vibration of the pantograph. Then, the second embodiment relates to a method of detecting a pantograph by comparing an image captured by a camera provided at the upper end of a railway with a pantograph template, and detecting pantograph vibration of an electric railway vehicle by using an image processing technique.

Hereinafter, the first embodiment is explained.

Fig. 1a and 1b are block diagrams sequentially showing the whole steps according to the first embodiment of the present invention, and fig. 2 shows a contact line dynamic pull-out value detecting apparatus 1 according to the first embodiment of the present invention. Referring to fig. 2, the contact line dynamic withdrawal value detection apparatus 1 includes a training unit 10, a detection unit 30, and a contact intersection calculation unit 50. The contact line dynamic pull-out value detection apparatus 1 can compare with the reference image brightness to detect the dynamic pull-out values of the pantograph 70 and the contact line 90 of the input image. The training section 10 may classify similar images according to different luminance conditions and obtain a plurality of reference images, respectively, and acquire the optimal feature vector 101. The training section 10 compares the feature vector generated from the selected reference image with the feature vector obtained from the input image, and can thereby specify in advance the pantograph template 80 representing the reference image to be compared. The detection unit 30 may detect the pantograph 70 and the contact line 90 present in the background image from the pantograph template 80. The contact intersection calculation unit 50 may detect the contact surface 709 and the center point 707 from the pantograph template 80, and calculate the dynamic pullout values of the contact line 90 and the pantograph 70 using the contact surface 709 and the center point 707. The contact line dynamic pull-out value detection apparatus 1 can detect the dynamic pull-out value of an input image without being disturbed by brightness.

The training section 10 may be defined to match different luminance conditions to a predetermined number of reference images 80 according to the image acquisition conditions. The training unit 10 may calculate the feature vector 101 by digitizing the luminance, variance, correlation, and the like of the reference image, and classify the reference image 80 into different categories 103 based on the feature vector 101. The training section 10 may acquire a plurality of similar images for defining the reference image 80 using a camera and a sensor provided on the vehicle, and the acquired images may show different degrees of light, exposure values of lenses, focuses, and the like. For example, if sunlight is reflected onto the pantograph 70, a pantograph 70 image with high brightness is obtained, and in the case of a cloud, a relatively dark pantograph 70 may be shown. In this case, since the detection performance of the pantograph 70 may be greatly degraded, it is necessary to define a case in which the pantograph images thus obtained are significantly different, to first perform a process of collecting the reference image 80 by different kinds for the case. The training unit 10 collects reference images for each defined category, and when there is a difference in the feature vectors 101 even in the same category of reference images, a condition for distinguishing the category 103 is required, and a Support Vector Machine (SVM) executes this function. The support vector machine may store image feature information such as an average value of brightness, contrast, and the like of the reference image 80, which may be quantified, in the form of the feature vector 101. The feature vector 101 stores information that can be digitized in vector form, and generally the stored information may be different depending on the luminance condition.

Fig. 3 shows a process of calculating a feature vector 101 and a step of classifying a reference image into different categories 103 with reference to the feature vector 101 according to a first embodiment of the present invention. Referring to fig. 3, the training unit 10 may minimize the classification error of the class 103 by using a support vector machine. The training section 10 may perform a linear classification process that performs training of classifying the reference image. The training unit 10 may extract features from reference images including the plurality of pantographs 70 and the contact line 90 acquired under various luminance conditions, and digitize them according to the luminance conditions. The training section 10 may provide a numerical brightness condition and provide information to enable the detection section 30 to determine the pantograph template 80 from the numerical value.

The training unit 10 may digitize the average value of the luminance in the reference image, the luminance of the contact line 90, and the contrast with the pantograph 70 to store them as the feature vector 101. The feature vector 101 may be digitized so as to represent one reference image, and the training section 101 may refer to the feature vector 101 to classify the reference image into different categories 103.

Fig. 4 is a schematic diagram illustrating reference images classified into different categories 103 according to the first embodiment of the present invention. Referring to fig. 4, the support vector machine calculates a luminance condition from a reference image and stores it as a feature vector 101, and classifies the reference image into different categories 103 with reference to the stored feature vector 101. The support vector machine classifies a plurality of reference images measured by a camera into categories 103 such as bright, dark, tunnel sections, background interference, and the like. The category 103 may be subdivided and set according to the luminance condition. The training section 10 may repeatedly perform the above-described classification process each time the reference image is obtained. The training section 10 can perform classification of the category 103 by the feature vector 101, and therefore the process of calculating the feature vector 101 is important.

Fig. 5 is a schematic diagram of a process of resetting the class 103 according to the first embodiment of the present invention. Referring to fig. 5, when the difference value between the average brightness of the pantograph region 104 and the average brightness of the class 103 in the updated feature vector 101 exceeds the set value, the updated feature vector 101 is calculated to perform a process of resetting the class 103. In an embodiment of the invention, the average luminance difference between the classes 103 is calculated, more than half of the difference is set as the reset threshold 105, and the classes 103 are reset when the luminance difference of the current image exceeds the corresponding reset threshold 105.

Fig. 6 is a simplified two-dimensional schematic diagram of the feature vectors 101 for each class 103 according to the first embodiment of the present invention. When the average brightness of the pantograph region 104 and the average brightness of the lower region of the pantograph in the image obtained in the tunnel interior and the normal image with clear sky as the background are plotted, the two-dimensional feature vector 101 is formed.

The training unit 10 of the present invention can set a determination boundary for determining which category 103 the feature vector 101 for determining the category 103 corresponds to using a support vector machine algorithm. However, in the present invention, the method of obtaining the decision boundary for the feature vector 101 of the category 103 is not limited to the support vector machine.

The training section 10 may perform the following steps: selecting categories 103 of normal conditions, tunnel conditions, background interference conditions, arc discharge occurrence conditions and the like required for classification to collect reference images; digitizing features such as luminance, correlation, and variance of the input image of each region to constitute a feature vector 101; resetting the category 103 to update the category 103; comparing the feature vector of the input image with the feature vector 101 of the current class to determine a class 103 that is most suitable for the input image; pantograph templates 80 corresponding to the determined category 103 are selected. The detection unit 30 may acquire the position of the pantograph 70 using the reference pantograph image 80 determined by the training unit 10 and calculate the dynamic pullout value of the contact line 90 based on the position of the pantograph.

Fig. 7 is a schematic diagram of the detection section 30 according to the first embodiment of the present invention detecting the pantograph 70 from the pantograph template 80. Referring to fig. 7, the detection unit 30 may detect the pantograph 70 from the pantograph template 80 determined by the training unit 10. The detection unit 30 extracts the pantograph 70 from the pantograph template 80 by using the following [ equation 1 ].

[ mathematical formula 1]

Where I is the input image, T is the pantograph template image, w is the input vector, X is the coordinates within the pantograph template image, and p is the linear transformation parameter of the input vector.

The detection section 30 can determine the optimal pantograph template 80 through the recognition process and perform the linear transformation w (x, p) of the input image. The detection unit 30 can obtain the position with the highest similarity from the pantograph template 80 by this calculation.

Fig. 8 is a schematic diagram of detecting a contact line 90 by fusing hough transform as a method of detecting straight lines and binarization as a method of distinguishing a background from a panorama according to the first embodiment of the present invention. Referring to fig. 8, the detection part 30 may perform the following steps: carrying out binarization on an input image; and limiting the brightness, height, width, area and angle of the contact line 90 and comparing the binary values to the pantograph template 80. The detection unit 30 may extract the brightness of the input image when detecting the pantograph 70 and the contact line 90, and use [ equation 1] to adopt a template matching technique for comparison with the pantograph template 80.

Fig. 9a shows a correctly detected pantograph 70 image according to the first embodiment of the present invention, and fig. 9b shows an incorrectly detected pantograph 70 image according to the first embodiment of the present invention. Referring to fig. 9a, in the case where the luminance conditions are similar, the detection section 30 can detect almost the same actual position and detection position. In this case, the dynamic pull-out value of the contact line 90 can be correctly calculated. Referring to fig. 9b, when the brightness condition is not satisfied, the detection section 30 detects different actual positions and detection positions, and thus may incorrectly calculate a dynamic pullout value. It can therefore be seen that in order to find the correct pantograph 70 and contact wire 90, the reference pantograph template 80 needs to be correctly set.

The contact intersection calculation unit 50 may calculate the dynamic pullout value of the contact line 90 using the detected pantograph 70 and the contact line 90. The contact intersection calculation unit 50 may perform the following steps: detecting a contact surface 709 corresponding to a level at which the detected contact line 90 and the pantograph 70 come into contact; detecting a center point 707 of the pantograph 70 from the contact surface 709; and the intersection of contact line 90 and contact face 709 is obtained. The contact intersection calculation unit 50 may calculate a dynamic pullout value of the intersection with the center point 707.

The step of detecting the contact surface 709 by the contact intersection calculation unit 50 may include the steps of: obtaining a straight line of the pantograph 70 from the input image using hough transform; and presume the position where the straight line makes contact with the contact line 90 after rising as the contact surface 709. That is, since the pantograph 70 is in close contact with the contact line 90 suspended in the air by air pressure, data similar to the actual contact position can be obtained by finding the position where the horizontal plane of the pantograph 70 and the contact line 90 intersect with each other by the binary image. In this case, a gap is generated between the pantograph 70 and the contact line 90 due to the vibration, but the difference of the gap is very small, and therefore even the approximation is included in the error range. Therefore, the contact intersection calculation section 50 may detect the contact surface 709, the center point 707, and the intersection, and calculate the distance from the center point 707 to the intersection to calculate the dynamic pullout value of the contact line 90. The distance detected in pixel units is converted to mm units using a proportional equation for the length of the pantograph 70 at the detected intersection point, and an accurate value is measured.

Although the measurement can be performed using the dynamic pullout value detection apparatus 1 for a contact wire as described above, there is also a method of using an attachment mark to simplify pantograph detection.

Fig. 10a shows a process of detecting a contact surface and a center point using an attached mark according to a first embodiment of the present invention. Fig. 10b shows an attached label according to the first embodiment of the present invention. Referring to fig. 10a, additional markers such as fig. 10b may be attached on the pantograph 70. In this case, the detection of the pantograph 70 and the contact surface 709 detection method can be simplified. In contrast to the detection section 30 needing to find a detection area corresponding to the image size of the pantograph 70 in the entire image as described above, in the case of the attached marker, only a part of the attached marker needs to be detected, and therefore the amount of calculation can be greatly reduced.

The step of detecting the contact surface 709 by the detecting portion 30 may include the steps of: obtaining a straight line from a mark attached to the upper end of the pantograph 70; the position where the straight line makes contact with the contact line 90 after rising is estimated as a contact surface 709. The step of detecting the center point 707 by the detecting section 30 may include the steps of: extracting a mark attached in the middle from marks attached in a straight line to the upper end of the pantograph 70; and the position of the marker attached in the middle is estimated as the center point 707, so the amount of calculation can be reduced.

In an embodiment of the present invention, 3 markers are attached to the pantograph 70. The first marker 701 may be attached to the left side of the pantograph 70, the second marker 703 may be attached to the middle of the pantograph 70, and the third marker 705 may be attached to the right side of the pantograph 70. When detecting three marks provided in the horizontal direction, the detection unit 30 linearly connects the three detected positions and estimates the contact position as the contact surface 709. The detection unit 30 may estimate the position of the mark located in the middle as the center point 707. Therefore, the step of estimating the contact surface 709 and the center point 707 is reduced by extracting the mark, and the amount of calculation can be reduced.

The second embodiment is explained below.

Fig. 11 is a configuration diagram of a vibration detection apparatus of a pantograph of an electric railway vehicle according to a second embodiment of the present invention, fig. 12 is a diagram for explaining a method of detecting a pantograph by each image frame according to the second embodiment of the present invention, fig. 13 is a diagram for explaining a method of measuring an angle between a horizontal line and a sliding plate in a detected pantograph according to the second embodiment of the present invention, fig. 14 is a diagram for explaining a method of calculating a vibration frequency of a pantograph based on an angle according to the second embodiment of the present invention, and fig. 15a and 15b are diagrams for explaining a method of calculating a vibration intensity of a pantograph based on an angle size according to the second embodiment of the present invention.

Referring to fig. 11, the vibration detection device 11 of the pantograph of the electric railway vehicle may include a communication module 100, a memory 200, and a processor 300.

The communication module 100 receives a captured image of the pantograph.

The communication module 100 receives a captured image of the pantograph by performing data communication with a camera that captures an image of the pantograph. In this case, the camera may be provided at an upper portion of the electric railway vehicle to ensure an image of the pantograph. Such a camera may be set as a high-speed camera or a general camera, and the higher the resolution, the more precisely the pantograph can be detected.

The memory 200 stores a vibration detection program of the pantograph.

Among them, the memory 200 is a generic name of a nonvolatile storage device (such as a flash memory, SSD) which continuously holds stored information even if power is not supplied and a volatile storage device (such as a DRAM, SRAM) which requires power to hold stored information.

The processor 300 executes a vibration detection program of the pantograph.

The processor 300 may compare the pantograph template and the photographed image of the pantograph received through the communication module by executing a program to detect the pantograph by each image frame.

Referring to fig. 12, for the pantograph template, in order to accurately detect the pantograph from a pantograph captured image which may differ in brightness depending on the light intensity, the exposure value of the lens, the focus, and the like, a plurality of pantograph templates which differ in brightness and contrast may be designated in advance.

Therefore, by matching the pantograph captured image having different image brightness depending on weather, tunnel conditions, presence or absence of an obstacle, and the like with the pantograph template having the highest similarity among the plurality of pantograph templates, it is possible to accurately detect the pantograph for each image frame.

Illustratively, as shown in fig. 3 (a), the processor 300 may accurately detect the pantograph 310 by each image frame by matching the pantograph captured image with a pantograph template having the highest similarity among a plurality of pantograph templates.

Referring to fig. 13 and 14, the processor 300 may detect a center point 326 of the slide plate from the detected pantograph 310 by executing a program, and measure an angle between a horizontal line 330 extending from the center point 325 and an extension line 320 of the slide plate, and calculate a vibration frequency of the pantograph based on the angle calculated for each frame.

The processor 300 may detect a straight line corresponding to the slide by hough-transforming the photographed image. In other words, the processor 300 may obtain a straight line corresponding to the pantograph 310 by hough-transforming each frame of the captured image. At this time, the straight line of the upper end of the pantograph 310 obtained by hough transform is estimated as a sled, and a straight line corresponding to the sled can be detected.

Illustratively, as shown in (b) and (c) of fig. 13, the processor 300 may detect a center point 325 of the slide board and a horizontal line 330 of the slide board and an extension line 320 of the slide board extending from the center point 325 of the slide board from the detected slide board.

Next, the processor 300 may measure the angle between the intersecting horizontal line 330 and the extension line 320 at the center point 325. At this time, the angle between the horizontal line 330 and the extension line 320 may be calculated for each frame of the photographed image, respectively.

Referring to fig. 14, the processor 300 may calculate a vibration frequency of the pantograph based on the angle calculated by each frame. Wherein the vibration frequency may be calculated based on the number of pantograph frames at which the measurement angle crosses zero divided by the total number of pantograph frames (frames per second).

As shown in fig. 14, the processor 300 arranges the angles between the horizontal line 300 and the extended line 320 of the slide board measured in each frame of the photographed image received through the communication module 100 in order of the image frame number, so that the position where the zero crossing of the measured angle occurs can be detected. Further, the processor 300 may calculate the total number of frames (frames per second) of images captured during 1 second and the zero-crossing number of frames per second where the position where the zero-crossing of the angle occurs is detected. The processor 300 may calculate a value obtained by dividing the calculated zero-crossing number of frames by 2 by the number of frames per second as a frequency.

Referring to fig. 15a and 15b, the processor 300 may calculate the vibration intensity of the pantograph based on the angle magnitude.

As shown in fig. 15a, the processor 300 may arrange the amount of change in the size of the angle between the horizontal line 300 and the extended line 320 of the slide board measured in each frame of the photographed image received through the communication module 100 in order of the image frame number, thereby detecting the vibration intensity waveform.

For example, describing a method of calculating the vibration intensity according to the angle size with reference to fig. 15b, the processor 300 may detect the pantograph 310 separated from the background by each image frame. At this time, only the pantograph 310 can be normally detected by the binary image technique used in the background of the separate image. Next, a lower end line 320a of the slide plate extending in the lower end surface of the slide plate based on the center point 325 of the slide plate is detected, and the slide plate is tilted by rotating in the detected lower end line 320a of the slide plate (a slide plate angle shown in fig. 13), whereby a vibration intensity reference line 330a parallel to a horizontal line 330 of the slide plate can be detected. Further, the processor 300 may measure the left and right displacements of the lower end line 320a and the vibration intensity reference line 330a of the slide. Here, the slide board is a rigid body with almost no bending, and thus the right side descends when the left side ascends, so that the magnitude of the vertical displacement of the left and right sides of the slide board can be the same.

Therefore, as shown in fig. 15b, the vibration intensity can be calculated by the difference between the left-side displacement and the right-side displacement of the slide board. At this time, references of the left and right sides of the skateboard may be obtained by a template matching technique, but are not limited thereto. Further, the processor 300 may calculate a difference between the left and right side references of the slide board in mm using the scale information of each image pixel (mm), and convert the difference between the left and right side displacements in mm calculated by each image frame into a vibration intensity in mm.

Further, a method of calculating the vibration frequency of the pantograph based on the left-side displacement and the right-side displacement calculated by each frame is explained below.

Fig. 16 is a diagram for explaining a method of measuring a horizontal line and left and right displacements of the slide plate according to the second embodiment of the present invention, and fig. 17 is a diagram for explaining a method of calculating a frequency based on an average value of the left and right displacements according to the second embodiment of the present invention.

Referring to fig. 16, the processor 300 detects a center point 325 of a slide plate in a pantograph 310 detected per each image frame, and may measure left and right displacements of a horizontal line 330 extending from the center point 325 and an extension line 320 of the slide plate.

For example, referring to fig. 16 and 17, the processor 300 may detect a center point 325 of the slide board and a horizontal line 330 of the slide board and an extension line 320 of the slide board extending from the center point 325 of the slide board from the detected slide board. In addition, the processor 300 may measure the left and right displacements of the crossing horizontal line 330 and the sled extension line 320 from the center point 325. At this time, the left and right displacements of the horizontal line 330 and the extension line 320 may be calculated for each frame of the photographed image, respectively.

Referring to fig. 17, the processor 300 may calculate a vibration frequency of the pantograph 310 based on the displacement calculated by each frame. Wherein the vibration frequency may be calculated based on an average of the measured respective left and right displacements.

As shown in fig. 17, the processor 300 may measure the left side up-down displacement and the right side up-down displacement of the slide board for each frame within 1 second of the photographed image received through the communication module 100. Further, the processor 300 may perform fourier transform on each average up-down displacement of the left and right sides thus measured to calculate the frequency.

For example, in the case of calculating the frequency using a fast fourier transform algorithm that transforms the obtained signal into frequency components and the intensities of the respective components, the frequency may be calculated by reflecting the sampling rate determined from the image acquisition frequency to a fast fourier transform.

Fig. 18 is a flowchart for explaining a method of detecting vibration of a pantograph of an electric railway vehicle according to a second embodiment of the present invention. The following description will be omitted for the structure that performs the same function as in the structure shown in fig. 11 to 17 described above.

First, a pantograph template and an image photographed with a camera are compared to detect a pantograph by each image frame (step S110).

Next, the center point of the slide plate is detected in the pantograph detected for each image frame, and the angle between the horizontal line extending from the center point and the extension line of the slide plate is measured (step S120).

Then, the vibration frequency of the pantograph is calculated based on the angle calculated for each frame (step S130).

The pantograph detecting step S110 may compare the pantograph template and the image photographed with the camera to detect the pantograph by each image frame. For example, since the camera is provided at the upper portion of the electric railway vehicle to capture an image of the pantograph provided at the upper portion of the electric railway vehicle, the brightness of the image captured according to the external environment may be different. That is, the brightness of the pantograph image captured in the tunnel condition, the presence or absence of an obstacle, or the like may be different depending on the weather. Therefore, in order to accurately detect a pantograph without being affected by a case where images of respective frames are different, a plurality of pantograph templates having various luminances and contrasts may be specified in advance. Therefore, when detecting a pantograph for each frame, the pantograph can be detected with high accuracy by matching a pantograph captured image with a pantograph template having the highest similarity between brightness and contrast among a plurality of pantograph templates.

The step S110 of detecting the pantograph may detect a straight line corresponding to the sled by performing hough transform on the photographed image.

For example, the step S110 of detecting the pantograph may obtain a straight line corresponding to the pantograph by performing hough transform on the pantograph detected by each image frame. Then, a straight line corresponding to the slide plate can be detected by estimating a straight line at the upper end of the pantograph as the slide plate.

The step S120 of measuring an angle may detect a center point of a straight line of the skateboard detected by hough transform. Next, a horizontal line extending from the detected center point may be detected. Then, the angle between the horizontal line of intersection and the extension line of the slide plate can be measured at the center point.

The step S130 of calculating the vibration frequency may calculate the frequency based on the number of pantograph frames in which the measurement angle crosses zero divided by the total number of pantograph frames (frames per second).

The step S130 of calculating the vibration frequency may arrange the angles between the horizontal line and the slide plate measured in each frame of the image photographed within 1 second in order of the image frame number, so that the position where the zero crossing of the measured angle occurs may be detected. Next, the total number of frames (frames per second) of images captured during 1 second and the number of zero-crossing frames in which the position where the zero-crossing of the angle occurs is detected in the frames per second can be calculated. Then, the frequency may be calculated based on a value obtained by dividing the zero-crossing frame number by 2 by the total number of frames (frames per second) of images captured during 1 second.

The vibration intensity of the pantograph may be calculated based on the magnitude of the angle after the step S130 of calculating the vibration frequency.

The step of calculating the vibration intensity may arrange the amount of change in the size of the angle between the horizontal line and the slide plate measured in each frame of the image photographed within 1 second in order of the image frame number, thereby detecting the vibration intensity waveform. The vibration intensity can be calculated from the waveform.

In addition, in step S130 of calculating the vibration frequency, the vibration frequency of the pantograph may be calculated from the left and right displacements calculated for each frame.

Before the step S130 of calculating the vibration frequency, a center point of the slide plate is detected from the pantograph detected for each image frame, and left and right displacements of a horizontal line extending from the center point and an extension line of the slide plate are measured, and in the step S130 of calculating the vibration frequency, the vibration frequency of the pantograph may be calculated based on the displacements calculated for each frame.

For example, before the step S130 of calculating the vibration frequency, the detected extension line of the slide plate and the center point of the slide plate may be detected by hough transform for each frame of the photographed image. Next, a horizontal line extending from the detected center point may be detected. Then, the horizontal line of intersection and the left and right displacements of the sled from the center point can be measured.

The step S130 of calculating the vibration frequency may calculate the frequency based on the average value of the respective left and right displacements thus measured.

Illustratively, the step S130 of calculating the vibration frequency may calculate the frequency by fourier-transforming each average up-down displacement of the left and right sides measured by each frame of the photographed image.

The vibration information detection device of the pantograph-based electric railway vehicle described above may also be realized in the form of a recording medium containing instructions executable by a computer, such as program modules executed by the computer. Computer readable media can be any available media that can be accessed by the computer and includes both volatile and nonvolatile media, removable and non-removable media. Computer-readable media may also include computer storage media and communication media. Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media.

The foregoing description of the present invention is merely exemplary, and it will be understood by those skilled in the art that other embodiments may be easily modified without changing the technical idea or essential features of the present invention. The embodiments described above are therefore to be considered in all respects as illustrative and not restrictive. For example, each component described in a single form may be dispersed and implemented, and similarly, components described in a dispersed form may be implemented in a combined form.

It should be understood that the scope of the present invention is defined by the appended claims, not the above detailed description, and all modifications and variations derived from the meaning and scope of the claims and their equivalents are included in the scope of the present invention.

Claims

1. A method of detecting vibration information of an electric railway vehicle, comprising the steps of:

determining a pantograph template;

comparing the pantograph template with an input image obtained with a camera to detect a pantograph; and

detecting vibration information of the pantograph based on a change in a position of the pantograph detected by comparison between each frame of the input image and the pantograph template,

wherein the step of determining the pantograph template comprises the steps of: obtaining a reference image under different brightness conditions and matching the brightness conditions to set the reference image; and determining, in the set reference image, a pantograph template representing a reference image to be compared with the input image for comparison,

the step of detecting the pantograph comprises the steps of: detecting the pantograph and contact line from the pantograph template,

the step of detecting vibration information of the pantograph includes the steps of: calculating dynamic pull-out values for the contact line and the pantograph.

2. The method of detecting vibration information of an electric railway vehicle according to claim 1, wherein the step of setting the reference image includes the steps of:

the reference images are collected separately in different categories such as normal, background interference, arcing, tunneling.

3. The method of detecting vibration information of an electric railway vehicle according to claim 2, comprising: in a case where the class needs to be updated due to a difference in the input images, an average brightness of a pantograph region and a class of the input images is compared to reset a pantograph template that is a reference for detecting a pantograph in the input images.

4. The method of detecting vibration information of an electric railway vehicle according to any one of claims 1 to 3, wherein the step of determining the pantograph template comprises the steps of:

digitizing the brightness, variance and correlation characteristics of each region of the input image to calculate a characteristic vector of the input image; and

comparing the feature vector of the input image and the feature vector of the pantograph template to find the feature vector of the pantograph template which is the same as or most similar to the feature vector of the input image, to determine a target class and a target pantograph template.

5. The method of detecting vibration information of an electric railway vehicle according to any one of claims 1 to 3, wherein the step of detecting the pantograph and the contact line includes the steps of:

simultaneously carrying out binarization and Hough transformation of the image so as to digitize the input image; and

setting the brightness, height, width, area and angle of the contact line and comparing the binarized value with the pantograph template.

6. The method of detecting vibration information of an electric railway vehicle according to claim 5, wherein in the step of detecting the pantograph and the contact line, a template matching technique of extracting a luminance value of the input image to compare with the pantograph template is used to detect the pantograph.

7. The method of detecting vibration information of an electric railway vehicle according to any one of claims 1 to 3, wherein the step of calculating the dynamic pull-out value includes the steps of:

detecting a contact surface corresponding to a horizontal plane where the contact line and the pantograph come into contact;

detecting a center point of the pantograph from the contact surface; and

the intersection of the contact line and the contact surface is obtained,

wherein a dynamic pull-out value of the intersection point from the center point is calculated.

8. The method of detecting vibration information of an electric railway vehicle according to claim 7, wherein the step of detecting the contact surface includes the steps of:

obtaining a straight line corresponding to a contact surface between the pantograph and a contact line from the input image by using hough transform; and

the position where the straight line makes contact with the contact line after rising is estimated as a contact surface.

9. The method of detecting vibration information of an electric railway vehicle according to claim 7, wherein the step of detecting the contact surface includes the steps of:

detecting a plurality of markers attached in a straight line to an upper end of the pantograph to obtain the straight line; and

10. The method of detecting vibration information of an electric railway vehicle according to claim 9, wherein the step of detecting the center point comprises the steps of:

extracting a position of a mark attached to a middle from a plurality of marks arranged in a straight line at an upper end of the pantograph; and

the position of the marker attached in the middle is estimated as the center point.

11. A method of detecting vibration information of an electric railway vehicle, comprising the steps of:

determining a pantograph template;

wherein the step of detecting the pantograph comprises the steps of: comparing the pantograph template with the input image to detect a pantograph by each image frame,

the step of detecting vibration information of the pantograph includes the steps of: detecting a center point of a slide plate in the detected pantograph by each image frame, and measuring an angle between a horizontal line extending from the center point and an extension line of the slide plate; and calculating a vibration frequency of the pantograph based on the change in the angle calculated for each image frame.

12. The method of detecting vibration information of an electric railway vehicle according to claim 11, wherein the step of detecting vibration information of the pantograph further comprises the steps of: calculating the vibration intensity of the pantograph based on the size of the angle.

13. The method of detecting vibration information of an electric railway vehicle according to claim 11, wherein the step of detecting the pantograph comprises the steps of: a straight line corresponding to the sled is detected by subjecting the obtained image to Hough transform.

14. The method of detecting vibration information of an electric railway vehicle according to claim 11, wherein in the step of calculating the vibration frequency, the frequency is calculated based on a value obtained by dividing a pantograph frame number at which the measured angle zero-crosses by a pantograph total frame number.

15. The method of detecting vibration information of an electric railway vehicle according to claim 11, wherein before the step of calculating the vibration frequency, comprising the steps of: detecting a center point of a slide plate in the detected pantograph by each image frame, and measuring left and right displacements of a horizontal line extending from the center point and an extension line of the slide plate,

in the step of calculating the vibration frequency, the vibration frequency of the pantograph is calculated based on the displacement calculated for each image frame.

16. The method of detecting vibration information of an electric railway vehicle according to claim 15, wherein in the step of calculating the vibration frequency, the frequency is calculated based on an average value of the measured each of the left-side displacement and the right-side displacement.

17. An apparatus for detecting vibration information of an electric railway vehicle, comprising:

a communication module that receives a captured image of a pantograph;

a memory that stores a vibration information detection program of the pantograph; and

a processor that executes a vibration information detection program of the pantograph,

wherein the processor executes the program so as to execute the method of detecting vibration information of an electric railway vehicle according to claim 1 or 11.