CN112418323B - Railway wagon coupler knuckle pin fault detection method based on image processing - Google Patents
Railway wagon coupler knuckle pin fault detection method based on image processing Download PDFInfo
- Publication number
- CN112418323B CN112418323B CN202011333887.8A CN202011333887A CN112418323B CN 112418323 B CN112418323 B CN 112418323B CN 202011333887 A CN202011333887 A CN 202011333887A CN 112418323 B CN112418323 B CN 112418323B
- Authority
- CN
- China
- Prior art keywords
- image
- coupler knuckle
- coupler
- knuckle pin
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/752—Contour matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
A fault detection method for a coupler knuckle pin of a railway wagon based on image processing relates to fault detection for the coupler knuckle pin of the railway wagon; the method solves the problem that the accuracy of the detection result is low in a method for detecting whether the latch bolt is lost or broken by adopting a manual checking mode. According to the method, corresponding template images are called according to the coupler types of two adjacent carriages; preprocessing an image of a coupler combination area, and matching and correcting the obtained preprocessed image and the called template image to obtain an initial coupler knuckle pin identification area image; acquiring two images of the to-be-identified fault area of the coupler knuckle pin according to the outline characteristics of the image of the initial identification area of the coupler knuckle pin; judging whether a pin hole in each image of the to-be-identified fault area of the coupler knuckle pin exists, if so, judging the fault, if not, judging whether the coupler knuckle pin in the image of the to-be-identified fault area of the coupler knuckle pin is broken by utilizing an SVM separator model, and thus, completing fault detection. The method is used for detecting the failure of the latch bolt.
Description
Technical Field
The invention belongs to the field of fault detection of rail wagons, and particularly relates to fault detection of a coupler knuckle pin of a rail wagon.
Background
The coupler of the railway wagon is a vehicle part used for realizing coupling between a locomotive and a carriage or between the carriage and the carriage, transmitting traction force and impact force and keeping a certain distance between vehicles. The coupler comprises a coupler knuckle and a coupler body, and the coupler is assembled as follows: two side surfaces of the coupler knuckle are respectively provided with a pin hole which is connected with a coupler knuckle pin and assembled at the mounting hole of the coupler body, and the coupler knuckle can rotate around the coupler knuckle pin. When the carriages are connected and hung, and the car couplers corresponding to the carriages of different types are different, so that the coupler knuckle pin is frequently subjected to the action of pulling force, compression force and impact force in the running process of a train, and the coupler knuckle pin is cracked, falls off, broken and other faults in a long time. Once the faults occur in the running process of the train, the train is easily separated, the emergency braking of the train is forced, wheel set grooves are scratched and the like, and serious accidents such as the derailment and overturn of the train can be caused in serious cases. Therefore, inspection of the latch pin for failure is enhanced when the vehicle is inspected.
And for the failure that whether the coupler knuckle pin is lost or broken, the failure is generally overhauled in a manual checking mode at the present stage. Due to the fact that the influence of factors such as business quality, responsibility and labor intensity of operators is large in the process, conditions such as missing detection or operation simplification are prone to occurring. If the fault cannot be found in time, serious vehicle fault is easily caused; therefore, the above problems need to be solved.
Disclosure of Invention
The invention aims to solve the problem of low accuracy of a detection result of a method for detecting whether a coupler knuckle pin is lost or broken by adopting a manual checking mode, and provides a fault detection method for a coupler knuckle pin of a railway wagon based on image processing.
The method for detecting the fault of the coupler knuckle pin of the railway wagon based on image processing comprises the following steps of:
acquiring images of two adjacent carriages of a passing vehicle and images of a coupler combination area between the two adjacent carriages;
determining the coupler types of two adjacent compartments of the passing vehicle according to the images of the two adjacent compartments of the passing vehicle, and calling corresponding template images according to the coupler types of the two adjacent compartments;
preprocessing an image of a coupler coupling area between two adjacent carriages, and matching and correcting the preprocessed image and the called template image to obtain an initial coupler knuckle pin identification area image;
step four, obtaining the contour characteristics of the initial identification area image of the coupler knuckle pin according to the initial identification area image of the coupler knuckle pin; obtaining two images of the to-be-identified fault area of the coupler knuckle pin according to the outline characteristics of the image of the initial identification area of the coupler knuckle pin;
step five, judging whether a pin hole in the image of the to-be-identified fault area of each coupler knuckle pin exists or not, if so, taking the result as a detection result, executing step seven, and if not, executing step six;
judging whether the coupler knuckle pin in the image of the fault area to be identified of the coupler knuckle pin is broken or not by utilizing an SVM separator model, if the coupler knuckle pin is broken, taking the broken result of the coupler knuckle pin as a detection result, and executing a seventh step;
and step seven, sending the detection result to a remote terminal.
Preferably, in the second step, the implementation manner of determining the coupler type of the two adjacent cars of the passing vehicle according to the two adjacent car images of the passing vehicle includes:
the method comprises the steps of obtaining the car type and the car number of each car according to images of two adjacent cars of a passing car, determining the type of a coupler of the car according to the car type and the car number of the car, and finally obtaining the type of the coupler of the two adjacent cars of the passing car.
Preferably, in the third step, the implementation manner of preprocessing the image of the coupler coupling area between two adjacent cars is as follows:
and carrying out histogram equalization on images of a coupler combination area between two adjacent carriages.
Preferably, the step three of matching and correcting the obtained preprocessed image and the called template image to obtain the image of the initial identification area of the coupler knuckle pin includes:
step three, matching the preprocessed image with the called template image to obtain the positions of the areas where the two latch hooks are located in the preprocessed image;
step two, according to the corresponding relation between the positions of the areas where the two coupler knuckle pins are located in the preprocessed image and the acquired image of the coupler combination area between two adjacent carriages, intercepting the area image where the two coupler knuckle pins are located from the acquired image of the coupler combination area between the two adjacent carriages;
and thirdly, correcting the area images of the two coupler knuckle pins intercepted from the acquired images of the coupler combination area between the two adjacent carriages, wherein the corrected images are used as initial coupler knuckle pin identification area images.
Preferably, the step three, matching the preprocessed image with the called template image, and obtaining the positions of the areas where the two latch pins are located in the preprocessed image, includes:
and respectively extracting the feature points of the preprocessed image and the called template image by using an ORB algorithm, and performing feature matching on the feature points extracted from the two images so as to determine the positions of the areas where the two coupler knuckle pins are located in the preprocessed image.
Preferably, in the fourth step, the implementation manner of obtaining the contour feature of the image of the initial identification area of the coupler knuckle pin according to the image of the initial identification area of the coupler knuckle pin includes:
processing the initial identification area image of the latch bolt by a canny edge detection algorithm to obtain the position coordinates of each contour point in the initial identification area image of the latch bolt and the gradient direction corresponding to the contour point; and the position coordinates of all contour points in the initial coupler knuckle pin identification area image and the gradient directions of all contour points form the contour characteristics of the initial coupler knuckle pin identification area image.
Preferably, in the fourth step, the implementation manner of obtaining the images of the to-be-identified fault areas of the two latch pins according to the outline features of the image of the initial identification area of the latch pins includes:
fourthly, according to the position coordinates of each contour point in the initial identification area image of the coupler knuckle pin, the radius R of the coupler knuckle pin and the angle parameter
Obtaining the center coordinates of a circle on which each contour point is located;
step two, according to the center coordinates of the circle where each contour point is located and the position coordinates of the contour point, the normal direction theta of the contour point on the circle is obtained, the normal direction theta of the contour point is differed with the gradient direction of the contour point, and if the difference value is within the threshold range, the occurrence frequency of the center coordinates corresponding to the contour point is recorded;
step four, repeatedly executing the step four and the step four until all contour points in the initial identification area image of the coupler knuckle pin are traversed, so that the times of the appearance of circle center coordinates corresponding to all the contour points are obtained;
and fourthly, extracting two circle center coordinates with the largest occurrence frequency from the circle center coordinates corresponding to all the contour points, respectively using the two extracted circle center coordinates as the centers of the two latch pins, taking the two extracted circle center coordinates as the centers, intercepting the area image with the width and the length both being 1.5R, and using the two intercepted area images as the to-be-identified fault area images of the two latch pins, thereby completing the acquisition of the to-be-identified fault area images of the two latch pins.
Preferably, in the step four, the position coordinate, the knuckle pin radius R and the angle parameter of each contour point in the image of the initial identification area of the knuckle pin are used as the basis
The implementation mode of obtaining the center coordinates of the circle on the circle where each contour point is located comprises the following steps:
wherein x is the abscissa of the contour point, y is the ordinate of the contour point, and x0Abscissa as center of circle, y0A vertical coordinate as a circle center;
in the second step, the implementation manner of obtaining the normal direction θ of each contour point on the circle according to the center coordinate of the circle on which the contour point is located and the position coordinate of the contour point includes:
preferably, in the step five, the implementation manner of judging whether the pin hole in the image of the to-be-identified fault region of each coupler knuckle pin exists includes:
and calculating the difference between the gray average value in the area which takes the center of each coupler knuckle pin to-be-identified fault area image as the circle center and R as the radius and the gray average value of the rest area in the coupler knuckle pin to-be-identified fault area image, and if the gray difference value is greater than the gray threshold value, judging that the pin hole in the coupler knuckle pin to-be-identified fault area image exists.
Preferably, in the sixth step, the implementation manner of determining whether the coupler knuckle pin is broken in the image of the to-be-identified fault area of the coupler knuckle pin by using the SVM separator model includes:
and extracting a gray level co-occurrence matrix in an area with the center of the image of the to-be-identified fault area of the coupler knuckle pin as the circle center and the radius of 1.5R as the radius, and classifying and identifying the gray level co-occurrence matrix through an SVM separator model so as to determine whether the coupler knuckle pin in the image of the to-be-identified fault area of the coupler knuckle pin is broken or not.
The invention has the following beneficial effects: when the method is applied specifically, high-definition imaging equipment is arranged at the bottom of a railway to shoot a wagon passing through the equipment, an image is obtained, whether the coupler knuckle pin in the image is lost or broken is detected by using an image processing technology through the method for detecting the coupler knuckle pin fault of the railway wagon based on image processing, and then the image is uploaded to a network. For manual review. And the staff performs corresponding processing according to the image recognition result to ensure the safe operation of the locomotive.
1. According to the invention, the image automatic identification mode is used for replacing manual detection, so that the fault identification detection efficiency and accuracy are improved, and the labor cost is reduced.
2. The invention extracts possible fault areas firstly, namely: and if the pin holes do not exist in the image of the fault area to be identified, classification and identification are carried out through the SVM separator model, so that the fault identification efficiency can be effectively improved.
3. The invention adopts the mode of combining image processing and machine learning (namely, SVM separator model) to judge the fault, improves the accuracy of fault identification, avoids missing report and reduces false report.
Drawings
FIG. 1 is a flow chart of a railway wagon coupler knuckle pin fault detection method based on image processing according to the invention;
fig. 2 is a relative schematic diagram of an initial identification area image of the coupler knuckle pin and an image of a fault area to be identified of the coupler knuckle pin.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
Referring to fig. 1 and fig. 2, the embodiment of the method for detecting a coupler knuckle pin fault of a railway wagon based on image processing includes the following steps:
acquiring images of two adjacent carriages of a passing vehicle and images of a coupler combination area between the two adjacent carriages;
determining the coupler types of two adjacent compartments of the passing vehicle according to the images of the two adjacent compartments of the passing vehicle, and calling corresponding template images according to the coupler types of the two adjacent compartments;
preprocessing an image of a coupler coupling area between two adjacent carriages, and matching and correcting the preprocessed image and the called template image to obtain an initial coupler knuckle pin identification area image;
step four, obtaining the contour characteristics of the initial identification area image of the coupler knuckle pin according to the initial identification area image of the coupler knuckle pin; obtaining two images of the to-be-identified fault area of the coupler knuckle pin according to the outline characteristics of the image of the initial identification area of the coupler knuckle pin;
step five, judging whether a pin hole in the image of the to-be-identified fault area of each coupler knuckle pin exists or not, if so, taking the result as a detection result, executing step seven, and if not, executing step six;
judging whether the coupler knuckle pin in the image of the fault area to be identified of the coupler knuckle pin is broken or not by utilizing an SVM separator model, if the coupler knuckle pin is broken, taking the broken result of the coupler knuckle pin as a detection result, and executing a seventh step;
and step seven, sending the detection result to a remote terminal.
In the embodiment, the automatic image identification mode is used for replacing manual detection, so that the fault identification detection efficiency and accuracy are improved, and the labor cost is reduced; the invention extracts possible fault areas firstly, namely: the two coupler knuckle pins are used for classifying and recognizing the fault area image to be recognized, if no pin hole exists, the SVM separator model is used for classifying and recognizing, and therefore the fault recognition efficiency can be effectively improved; the invention adopts the mode of combining image processing and machine learning (namely, SVM separator model) to judge the fault, improves the accuracy of fault identification, avoids missing report and reduces false report.
In a specific application, the images of the coupler connection areas between two adjacent cars are different due to different coupler types of the cars, for example, the coupler connection part can be a 13-type coupler and a 13-type coupler, or a 17-type coupler and a 17-type coupler, or a 17-type coupler and a 13-type coupler, or a 13-type coupler and a 17-type coupler. Therefore, in the second step, the coupler types of the two adjacent compartments of the passing vehicle are determined according to the images of the two adjacent compartments of the passing vehicle, and the corresponding template images are called according to the coupler types of the two adjacent compartments, so that the matching progress can be accelerated, and the operation efficiency is improved.
The SVM separator model in this embodiment can be implemented by the prior art, and is a trained model.
And in the third step, the images of the coupler combination area between two adjacent carriages are preprocessed, so that the influence of illumination on the acquired images can be reduced.
And seventhly, the detection result is the result of the existence of the pin hole or the result of the breakage of the coupler knuckle pin, the result can be uploaded in a data transmission mode or an image mode, and when the result is uploaded in the image mode, the image is the image of the fault area to be identified of the coupler knuckle pin.
Further, in the second step, an implementation manner of determining coupler types of two adjacent cars of the passing vehicle according to the two adjacent car images of the passing vehicle includes:
the method comprises the steps of obtaining the car type and the car number of each car according to images of two adjacent cars of a passing car, determining the type of a coupler of the car according to the car type and the car number of the car, and finally obtaining the type of the coupler of the two adjacent cars of the passing car.
Furthermore, in the third step, the image of the coupler coupling area between two adjacent cars is preprocessed in the following manner:
and carrying out histogram equalization on images of a coupler combination area between two adjacent carriages.
In the preferred embodiment, histogram equalization is performed on the image of the coupler coupling area between two adjacent cars to further reduce the influence of illumination on the image.
Furthermore, in step three, the matching and correcting the obtained preprocessed image and the called template image to obtain the image of the initial identification area of the coupler knuckle pin includes:
step three, matching the preprocessed image with the called template image to obtain the positions of the areas where the two latch hooks are located in the preprocessed image;
step two, according to the corresponding relation between the positions of the areas where the two coupler knuckle pins are located in the preprocessed image and the acquired image of the coupler combination area between two adjacent carriages, intercepting the area image where the two coupler knuckle pins are located from the acquired image of the coupler combination area between the two adjacent carriages;
and thirdly, correcting the area images of the two coupler knuckle pins intercepted from the acquired images of the coupler combination area between the two adjacent carriages, wherein the corrected images are used as initial coupler knuckle pin identification area images.
In the preferred embodiment, a specific implementation manner for matching and correcting the preprocessed image and the called template image is provided, the whole process is simple, the matched image is corrected, and the image of the initial identification area of the coupler knuckle pin with higher accuracy is further obtained
Furthermore, the third step of matching the preprocessed image with the called template image to obtain the positions of the two areas where the coupler knuckle pins are located in the preprocessed image includes:
and respectively extracting the feature points of the preprocessed image and the called template image by using an ORB algorithm, and performing feature matching on the feature points extracted from the two images so as to determine the positions of the areas where the two coupler knuckle pins are located in the preprocessed image.
Furthermore, in the fourth step, the implementation manner of obtaining the contour feature of the image of the initial identification area of the coupler knuckle pin according to the image of the initial identification area of the coupler knuckle pin includes:
processing the initial identification area image of the latch bolt by a canny edge detection algorithm to obtain the position coordinates of each contour point in the initial identification area image of the latch bolt and the gradient direction corresponding to the contour point; and the position coordinates of all contour points in the initial coupler knuckle pin identification area image and the gradient directions of all contour points form the contour characteristics of the initial coupler knuckle pin identification area image.
Furthermore, in the fourth step, the implementation manner of obtaining the image of the to-be-identified fault area of the two latch pins according to the outline feature of the image of the initial identification area of the latch pins includes:
fourthly, according to the position coordinates of each contour point in the initial identification area image of the coupler knuckle pin, the radius R of the coupler knuckle pin and the angle parameter
Obtaining the center coordinates of a circle on which each contour point is located;
step two, according to the center coordinates of the circle where each contour point is located and the position coordinates of the contour point, the normal direction theta of the contour point on the circle is obtained, the normal direction theta of the contour point is differed with the gradient direction of the contour point, and if the difference value is within the threshold range, the occurrence frequency of the center coordinates corresponding to the contour point is recorded;
step four, repeatedly executing the step four and the step four until all contour points in the initial identification area image of the coupler knuckle pin are traversed, so that the times of the appearance of circle center coordinates corresponding to all the contour points are obtained;
and fourthly, extracting two circle center coordinates with the largest occurrence frequency from the circle center coordinates corresponding to all the contour points, respectively using the two extracted circle center coordinates as the centers of the two latch pins, taking the two extracted circle center coordinates as the centers, intercepting the area image with the width and the length both being 1.5R, and using the two intercepted area images as the to-be-identified fault area images of the two latch pins, thereby completing the acquisition of the to-be-identified fault area images of the two latch pins.
Furthermore, the angle parameters in the step four
Is a variable, and
the value range of (1) is 0-360 degrees, and the value step length is 2.
Furthermore, in the fourth step, according to the position coordinate of each contour point in the initial identification area image of the coupler knuckle pin, the radius R of the coupler knuckle pin and the angle parameter
The implementation mode of obtaining the center coordinates of the circle on the circle where each contour point is located comprises the following steps:
wherein x is a contour point sit-acrossScale, y is the ordinate of the contour point, x0Abscissa as center of circle, y0A vertical coordinate as a circle center;
in the second step, the implementation manner of obtaining the normal direction θ of each contour point on the circle according to the center coordinate of the circle on which the contour point is located and the position coordinate of the contour point includes:
furthermore, in the fifth step, the implementation manner of judging whether the pin hole in the image of the to-be-identified fault area of each coupler knuckle pin exists includes:
and calculating the difference between the gray average value in the area which takes the center of each coupler knuckle pin to-be-identified fault area image as the circle center and R as the radius and the gray average value of the rest area in the coupler knuckle pin to-be-identified fault area image, and if the gray difference value is greater than the gray threshold value, judging that the pin hole in the coupler knuckle pin to-be-identified fault area image exists.
In the preferred embodiment, in a specific application, the latch pin is usually a round hole with a black stroke at the original installation position after being lost or broken, so that the difference between the gray level average value in the area with the radius R in the central area of the screenshot and the surrounding gray level average value is calculated, when the difference value is greater than the threshold value, the pin hole exists, and at this time, the latch pin is lost, and the failure is determined.
Furthermore, in the sixth step, the implementation manner of judging whether the coupler knuckle pin is broken in the image of the fault area to be identified of the coupler knuckle pin by using the SVM separator model comprises the following steps:
and extracting a gray level co-occurrence matrix in an area with the center of the image of the to-be-identified fault area of the coupler knuckle pin as the circle center and the radius of 1.5R as the radius, and classifying and identifying the gray level co-occurrence matrix through an SVM separator model so as to determine whether the coupler knuckle pin in the image of the to-be-identified fault area of the coupler knuckle pin is broken or not.
Although the invention herein has been described with reference to particular embodiments, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present invention. It is therefore to be understood that numerous modifications may be made to the illustrative embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that features described in different dependent claims and herein may be combined in ways different from those described in the original claims. It is also to be understood that features described in connection with individual embodiments may be used in other described embodiments.
Claims (9)
1. The method for detecting the fault of the coupler knuckle pin of the railway wagon based on image processing comprises the following steps of:
acquiring images of two adjacent carriages of a passing vehicle and images of a coupler combination area between the two adjacent carriages;
determining the coupler types of two adjacent compartments of the passing vehicle according to the images of the two adjacent compartments of the passing vehicle, and calling corresponding template images according to the coupler types of the two adjacent compartments;
preprocessing an image of a coupler coupling area between two adjacent carriages, and matching and correcting the preprocessed image and the called template image to obtain an initial coupler knuckle pin identification area image;
step four, obtaining the contour characteristics of the initial identification area image of the coupler knuckle pin according to the initial identification area image of the coupler knuckle pin; obtaining two images of the to-be-identified fault area of the coupler knuckle pin according to the outline characteristics of the image of the initial identification area of the coupler knuckle pin;
step five, judging whether a pin hole in the image of the to-be-identified fault area of each coupler knuckle pin exists or not, if so, taking the result as a detection result, executing step seven, and if not, executing step six;
judging whether the coupler knuckle pin in the image of the fault area to be identified of the coupler knuckle pin is broken or not by utilizing an SVM separator model, if the coupler knuckle pin is broken, taking the broken result of the coupler knuckle pin as a detection result, and executing a seventh step;
step seven, sending the detection result to a remote terminal;
the method is characterized in that in the fourth step, according to the outline characteristics of the initial identification area image of the coupler knuckle pin, the implementation mode of obtaining the to-be-identified fault area images of the two coupler knuckle pins comprises the following steps:
fourthly, according to the position coordinates of each contour point in the initial identification area image of the coupler knuckle pin, the radius R of the coupler knuckle pin and the angle parameter
Obtaining the center coordinates of a circle on which each contour point is located;
the implementation manner of obtaining the center coordinates of the circle on which each contour point is located includes:
wherein x is the abscissa of the contour point, y is the ordinate of the contour point, and x0Abscissa as center of circle, y0A vertical coordinate as a circle center;
step two, according to the center coordinates of the circle where each contour point is located and the position coordinates of the contour point, the normal direction theta of the contour point on the circle is obtained, the normal direction theta of the contour point is differed with the gradient direction of the contour point, and if the difference value is within the threshold range, the occurrence frequency of the center coordinates corresponding to the contour point is recorded;
step four, repeatedly executing the step four and the step four until all contour points in the initial identification area image of the coupler knuckle pin are traversed, so that the times of the appearance of circle center coordinates corresponding to all the contour points are obtained;
and fourthly, extracting two circle center coordinates with the largest occurrence frequency from the circle center coordinates corresponding to all the contour points, wherein the two extracted circle center coordinates are respectively used as the centers of the two latch pins, the two extracted circle center coordinates are used as the centers, the area images with the width and the length both being 1.5R are intercepted, and the two intercepted area images are used as the area images of the two to-be-identified faults of the latch pins.
2. The method for detecting the coupler knuckle pin fault of the railway wagon based on the image processing as claimed in claim 1, wherein the implementation manner of determining the coupler type of two adjacent cars of the passing vehicle according to the two adjacent car images of the passing vehicle in the step two comprises:
the method comprises the steps of obtaining the car type and the car number of each car according to images of two adjacent cars of a passing car, determining the type of a coupler of the car according to the car type and the car number of the car, and finally obtaining the type of the coupler of the two adjacent cars of the passing car.
3. The method for detecting the fault of the coupler knuckle pin of the railway wagon based on the image processing as claimed in claim 1, wherein the preprocessing of the image of the coupler joint area between two adjacent carriages in the third step is realized by:
and carrying out histogram equalization on images of a coupler combination area between two adjacent carriages.
4. The method for detecting the fault of the coupler knuckle pin of the railway wagon based on the image processing as claimed in claim 1 or 3, wherein the step three of matching and correcting the obtained pre-processed image with the called template image so as to obtain the initial identification area image of the coupler knuckle pin comprises the following steps:
step three, matching the preprocessed image with the called template image to obtain the positions of the areas where the two latch hooks are located in the preprocessed image;
step two, according to the corresponding relation between the positions of the areas where the two coupler knuckle pins are located in the preprocessed image and the acquired image of the coupler combination area between two adjacent carriages, intercepting the area image where the two coupler knuckle pins are located from the acquired image of the coupler combination area between the two adjacent carriages;
and thirdly, correcting the area images of the two coupler knuckle pins intercepted from the acquired images of the coupler combination area between the two adjacent carriages, wherein the corrected images are used as initial coupler knuckle pin identification area images.
5. The method for detecting the fault of the coupler knuckle pin of the railway wagon based on the image processing as claimed in claim 4, wherein the third step is that the preprocessed image is matched with the called template image, and the implementation mode of obtaining the positions of the areas where the two coupler knuckle pins are located in the preprocessed image comprises the following steps:
and respectively extracting the feature points of the preprocessed image and the called template image by using an ORB algorithm, and performing feature matching on the feature points extracted from the two images so as to determine the positions of the areas where the two coupler knuckle pins are located in the preprocessed image.
6. The method for detecting the fault of the coupler knuckle pin of the railway wagon based on the image processing as claimed in claim 1, wherein the step four of obtaining the contour feature of the initial coupler knuckle pin identification area image according to the initial coupler knuckle pin identification area image comprises:
processing the initial identification area image of the latch bolt by a canny edge detection algorithm to obtain the position coordinates of each contour point in the initial identification area image of the latch bolt and the gradient direction corresponding to the contour point; and the position coordinates of all contour points in the initial coupler knuckle pin identification area image and the gradient directions of all contour points form the contour characteristics of the initial coupler knuckle pin identification area image.
7. The image processing-based rail wagon coupler knuckle pin fault detection method according to claim 1,
in the second step, the implementation manner of obtaining the normal direction θ of each contour point on the circle according to the center coordinate of the circle on which the contour point is located and the position coordinate of the contour point includes:
8. the method for detecting the faults of the coupler knuckle pins of the railway wagon based on the image processing as claimed in claim 1, wherein the implementation manner of judging whether pin holes exist in the to-be-identified fault area image of each coupler knuckle pin in the fifth step comprises the following steps:
and calculating the difference between the gray average value in the area which takes the center of each coupler knuckle pin to-be-identified fault area image as the circle center and R as the radius and the gray average value of the rest area in the coupler knuckle pin to-be-identified fault area image, and if the gray difference value is greater than the gray threshold value, judging that the pin hole in the coupler knuckle pin to-be-identified fault area image exists.
9. The method for detecting the fault of the coupler knuckle pin of the railway wagon based on the image processing as claimed in claim 1, wherein the implementation manner of judging whether the coupler knuckle pin is broken or not in the image of the fault area to be identified of the coupler knuckle pin by using the SVM separator model in the sixth step comprises the following steps:
and extracting a gray level co-occurrence matrix in an area with the center of the image of the to-be-identified fault area of the coupler knuckle pin as the circle center and the radius of 1.5R as the radius, and classifying and identifying the gray level co-occurrence matrix through an SVM separator model so as to determine whether the coupler knuckle pin in the image of the to-be-identified fault area of the coupler knuckle pin is broken or not.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011333887.8A CN112418323B (en) | 2020-11-24 | 2020-11-24 | Railway wagon coupler knuckle pin fault detection method based on image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011333887.8A CN112418323B (en) | 2020-11-24 | 2020-11-24 | Railway wagon coupler knuckle pin fault detection method based on image processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112418323A CN112418323A (en) | 2021-02-26 |
| CN112418323B true CN112418323B (en) | 2021-07-16 |
Family
ID=74842016
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011333887.8A Active CN112418323B (en) | 2020-11-24 | 2020-11-24 | Railway wagon coupler knuckle pin fault detection method based on image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112418323B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115049713B (en) * | 2022-08-11 | 2022-11-25 | 武汉中导光电设备有限公司 | Image registration method, device, equipment and readable storage medium |
| CN115973125B (en) * | 2023-02-15 | 2023-06-06 | 慧铁科技有限公司 | Method for processing faults of automatic derailment braking device of railway wagon |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101334837A (en) * | 2008-07-31 | 2008-12-31 | 重庆大学 | Multi-method integrated license plate image positioning method |
| CN103034850A (en) * | 2012-12-21 | 2013-04-10 | 湖北工业大学 | Trouble of moving freight car detection system (TFDS) block key loss fault automatic identification method |
| CN103295027A (en) * | 2013-05-17 | 2013-09-11 | 北京康拓红外技术股份有限公司 | Freight wagon blocking key missing fault identification method based on support vector machine |
| CN103544489A (en) * | 2013-11-12 | 2014-01-29 | 公安部第三研究所 | Device and method for locating automobile logo |
| CN109345554A (en) * | 2018-09-12 | 2019-02-15 | 南京农业大学 | A kind of adhesion mushroom vision in-situ measuring method based on RGB-D camera |
| CN110264440A (en) * | 2019-05-10 | 2019-09-20 | 湖南大学 | Extensive train displacement fault detection method and system based on deep learning |
| CN110595804A (en) * | 2019-10-09 | 2019-12-20 | 哈尔滨市科佳通用机电股份有限公司 | Intelligent fault detection system for railway wagon running part |
| CN110659552A (en) * | 2019-04-26 | 2020-01-07 | 南京理工大学 | Tramcar obstacle detection and alarm method |
| CN111079819A (en) * | 2019-12-12 | 2020-04-28 | 哈尔滨市科佳通用机电股份有限公司 | Method for judging state of coupler knuckle pin of railway wagon based on image recognition and deep learning |
| CN111216763A (en) * | 2020-03-26 | 2020-06-02 | 唐山昆铁科技有限公司 | Rail crawling observation device |
| CN111452840A (en) * | 2020-04-17 | 2020-07-28 | 华北理工大学 | Railway steel rail crawling displacement detection method based on monocular vision measurement technology |
| CN111080598B (en) * | 2019-12-12 | 2020-08-28 | 哈尔滨市科佳通用机电股份有限公司 | Bolt and nut missing detection method for coupler yoke key safety crane |
-
2020
- 2020-11-24 CN CN202011333887.8A patent/CN112418323B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101334837A (en) * | 2008-07-31 | 2008-12-31 | 重庆大学 | Multi-method integrated license plate image positioning method |
| CN103034850A (en) * | 2012-12-21 | 2013-04-10 | 湖北工业大学 | Trouble of moving freight car detection system (TFDS) block key loss fault automatic identification method |
| CN103295027A (en) * | 2013-05-17 | 2013-09-11 | 北京康拓红外技术股份有限公司 | Freight wagon blocking key missing fault identification method based on support vector machine |
| CN103544489A (en) * | 2013-11-12 | 2014-01-29 | 公安部第三研究所 | Device and method for locating automobile logo |
| CN109345554A (en) * | 2018-09-12 | 2019-02-15 | 南京农业大学 | A kind of adhesion mushroom vision in-situ measuring method based on RGB-D camera |
| CN110659552A (en) * | 2019-04-26 | 2020-01-07 | 南京理工大学 | Tramcar obstacle detection and alarm method |
| CN110264440A (en) * | 2019-05-10 | 2019-09-20 | 湖南大学 | Extensive train displacement fault detection method and system based on deep learning |
| CN110595804A (en) * | 2019-10-09 | 2019-12-20 | 哈尔滨市科佳通用机电股份有限公司 | Intelligent fault detection system for railway wagon running part |
| CN111079819A (en) * | 2019-12-12 | 2020-04-28 | 哈尔滨市科佳通用机电股份有限公司 | Method for judging state of coupler knuckle pin of railway wagon based on image recognition and deep learning |
| CN111080598B (en) * | 2019-12-12 | 2020-08-28 | 哈尔滨市科佳通用机电股份有限公司 | Bolt and nut missing detection method for coupler yoke key safety crane |
| CN111216763A (en) * | 2020-03-26 | 2020-06-02 | 唐山昆铁科技有限公司 | Rail crawling observation device |
| CN111452840A (en) * | 2020-04-17 | 2020-07-28 | 华北理工大学 | Railway steel rail crawling displacement detection method based on monocular vision measurement technology |
Non-Patent Citations (2)
| Title |
|---|
| Detecting Rails in Images from a Train-Mounted Thermal Camera using a Convolutional Neural Network;Magnus Wedberg;《https://liu.diva-portal.org/smash/get/diva2:1111486/FULLTEXT01.pdf》;20171231;第1-57页 * |
| ＴＦＤＳ故障动态图像识别系统的设计;李九灵 等;《湖北工业大学学报》;20131031;第２８卷(第5期);第8-10页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112418323A (en) | 2021-02-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111079819B (en) | Method for judging state of coupler knuckle pin of railway wagon based on image recognition and deep learning | |
| CN112418323B (en) | Railway wagon coupler knuckle pin fault detection method based on image processing | |
| CN111079747B (en) | Railway wagon bogie side frame fracture fault image identification method | |
| CN111652227B (en) | Method for detecting damage fault of bottom floor of railway wagon | |
| CN111080598B (en) | Bolt and nut missing detection method for coupler yoke key safety crane | |
| CN110008932B (en) | Vehicle violation line-pressing detection method based on computer vision | |
| CN110334750B (en) | Power transmission line iron tower bolt corrosion degree image classification and identification method | |
| CN111652295B (en) | Railway wagon coupler yoke key joist falling fault identification method | |
| CN103745224A (en) | Image-based railway contact net bird-nest abnormal condition detection method | |
| CN111091548B (en) | Railway wagon adapter dislocation fault image identification method and system based on deep learning | |
| CN111080607B (en) | Rolling bearing oil slinging fault detection method based on image recognition | |
| CN106485694A (en) | A kind of high ferro contact net double-jacket tube connector six-sided nut based on cascade classifier comes off defective mode detection method | |
| CN111862029A (en) | Fault detection method for bolt part of vertical shock absorber of railway motor train unit | |
| CN111832571B (en) | Automatic detection method for truck brake beam strut fault | |
| CN111626976A (en) | Steel rail surface damage detection method based on pulse coupling neural network | |
| CN114820626B (en) | Intelligent detection method for automobile front face part configuration | |
| CN111080604B (en) | Image identification method for breakage fault of hook lifting rod of railway wagon | |
| CN115527170A (en) | Method and system for identifying closing fault of door stopper handle of automatic freight car derailing brake device | |
| CN112699794B (en) | Method for identifying dislocation fault images of middle rubber and upper and lower floor plates of wagon axle box rubber pad | |
| CN115272287B (en) | Fault detection method, medium and system for rail wagon buffer and slave plate | |
| CN112907524B (en) | Method for detecting fault of fire-proof plate of rail wagon based on image processing | |
| CN116092003A (en) | Automatic identification method for preventing lifting of collection card based on multi-view visual detection | |
| CN109978879B (en) | Box corner in-groove state detection method based on railway wagon loading video monitoring | |
| CN113859309B (en) | Car coupler state monitoring system | |
| Yang et al. | Research of computer vision fault recognition algorithm of center plate bolts of train |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |