JP4779770B2 - Trolley wire wear measuring device by image processing - Google Patents

Description

  The present invention relates to a trolley wire wear measuring apparatus using image processing. More specifically, the present invention relates to trolley wire wear measurement by image processing, and more particularly to an apparatus for measuring the width of a worn portion of a trolley wire.

The trolley line that supplies power to the electric railway vehicle comes into contact with the current collector every time the vehicle passes.
For this reason, the trolley wire gradually wears while the electric railway vehicle is operated, and if it is not replaced, it eventually breaks and causes an accident.
Therefore, the trolley wire has a wear limit, and the trolley wire is replaced based on the wear limit to ensure the safety of the electric railway vehicle.

There are two methods for measuring the wear of the trolley wire: a method of directly measuring the thickness of the trolley wire, and a method of calculating the width of the wear portion of the trolley wire and converting the thickness of the trolley wire from the wear width. .
As a method of directly measuring the thickness of the trolley wire, first, there is a method of measuring the thickness of the trolley wire using a ruler such as a caliper.

This is a method in which the thickness of the trolley wire portion that the operator wants to measure is manually measured using a ruler such as a vernier caliper, and the thickness of the trolley wire to be measured can be reliably obtained.
On the other hand, the measurement is laborious and cannot be automated, so it is difficult to measure a long distance section.

Another method for directly measuring the thickness of the trolley wire is to use an optical sensor.
This is done by pressing the rotating roller against the trolley wire, and measuring the amount of light received at the trolley wire portion sandwiched between the devices by means of a laser irradiation device and a light receiving device attached to the roller base so as to sandwich the trolley wire. The thickness of the trolley wire is converted.

  Although this method can continuously measure the thickness of the trolley wire, it is necessary to operate at low speed because it involves contact with the trolley wire. It is impossible to operate in the part where structures that collide with sensors such as air section, anchor, etc. exist, and it is necessary to separate the device from the measurement position so that it does not collide in the place where those existing structures exist .

As a method of measuring the width of the trolley wire wear portion, there is a method of measuring the trolley wire wear portion by irradiating a sodium lamp or laser light (Non-Patent Document 1).
This uses the fact that the lower part of the trolley wire has a round gourd shape, and the width of the cut portion becomes wider as the trolley wire is flattened due to wear, and the thickness of the trolley wire at that location is determined from the wear width. Convert.

  To measure the trolley wire wear part, the position of the light receiving part with the line sensor is precisely adjusted so that the reflected light from the trolley wire wear part is received by regular reflection when irradiated from a light source such as a sodium lamp or laser light. The trolley wire wear part is made into a whiteout state by imaging strong light due to regular reflection, and the width of the trolley wire wear part is obtained from the width of the whiteout part receiving the strong light. This method is non-contact and can be operated at high speed.

  However, it is easily affected by noise such as clamps sandwiching the trolley wire and structures reflected in the background, and there is no way to check the wrong measurement result due to some noise, which causes problems as trolley wire wear. In the end, it is necessary to confirm by using a method that directly measures the thickness of the trolley wire, and it is necessary to precisely adjust the position of the light source and light receiving device so that it receives specularly reflected light. There is.

On the other hand, in the “trolley wire wear measuring device by image processing (Japanese Patent Application No. 2005-06879 6 )” (hereinafter referred to as “prior application invention”), the line sensor image is not contacted by image processing for a short time. Can measure long distances.
However, since the background sky on the line sensor is dark as an imaging condition in order to emphasize the trolley wire wear part using the reflected light of illumination, it is only at night when the trolley line wear part is imaged brightly.

  Trolley wire wear can be measured by directly measuring the thickness of the trolley wire using a ruler such as a caliper, directly measuring the thickness of the trolley wire using an optical sensor, and setting the width of the trolley wire wear portion to a sodium lamp. There are a method of measuring by irradiating a laser beam and converting the thickness of the trolley wire from the worn portion, and a trolley wire wear measuring method by image processing as in the prior invention, and has the following problems.

(1) In the case of the method of directly measuring the thickness of the trolley wire using a ruler such as a caliper, since the measurement is performed manually by an operator, the measurement is troublesome and cannot be automated. It is difficult to measure between.

(2) In the case of a method of directly measuring the thickness of the trolley wire using an optical sensor, it is necessary to operate at a low speed because it involves contact between the rotating roller and the trolley wire. In addition, since the trolley wire is sandwiched between the sensors, it cannot be used in areas where structures that collide with sensors such as points, air sections, anchors, etc. cannot be used. It is necessary to move the device away from the measurement position so that it does not collide.

(3) In the method of measuring the width of the trolley wire wear portion by irradiating a sodium lamp or laser light and converting the wear width into the thickness of the trolley wire, first, prepare special illumination light such as sodium lamp or laser light In particular, when laser light is used, it is necessary to consider the influence on the human body, so care must be taken in handling. Next, since it is easily affected by noise such as clamps sandwiching the trolley wire and structures reflected in the background, there is no way to confirm the wrong measurement result due to some noise, so trolley wire wear Even if there is a problem, there is no means for confirming the cause of the problem, and eventually, it takes time and labor to directly measure the thickness of the wire. In addition, it is necessary to precisely adjust the positions of the light source and the light receiving device so as to receive specularly reflected light.

(4) In the case of the trolley wire wear measurement method by image processing as in the prior invention, the background sky on the line sensor as an imaging condition in order to emphasize the trolley wire wear portion using the reflected light of illumination. This is limited to nighttime when the trolley wire wear part is brightly imaged.

  The apparatus for measuring trolley wire wear according to claim 1 of the present invention for solving the above-described problem is a means for creating a line sensor image using a line sensor and storing it as an input image, and for the entire line sensor image. On the screen using the binarization processing means, the trolley line height parameter input from the outside with respect to the binarized line sensor image, and the actual trolley line thickness parameter. Means for performing reversal processing for calculating the thickness of the trolley line and reversing a white area larger than that to a black area, and binarization processing and noise removal processing for the line sensor image subjected to the reversal processing Means for detecting the edge of the trolley wire wear portion, with the white region in the line sensor image subjected to binarization processing, inversion processing and noise removal processing as a trolley wire wear portion, Characterized in that it comprises means from the issued the edge performing the calculation of the width of the trolley wire worn portion.

  A trolley wire wear measuring apparatus by image processing according to claim 2 of the present invention for solving the above-mentioned problems is provided with means for creating a line sensor image using a line sensor and storing it as an input image, and for the entire line sensor image. And a means for performing discriminant analysis binarization processing, and a trolley line height parameter and an actual trolley line thickness parameter input from the outside with respect to the line sensor image subjected to the discriminant analysis binarization process. Means for calculating the thickness of the trolley line on the screen and inverting the white area larger than that to the black area, and the line sensor image subjected to the discriminant analysis binarization process and the inverting process. A means for performing a noise removal process on the line sensor image, and a white area in the line sensor image subjected to the discriminant analysis binarization process, the reversal process, and the noise removal process as a trolley line wear part. It means for detecting the edge of 耗部, characterized in that from said detected edge comprises means for performing a calculation of the width of the trolley wire worn portion.

According to the trolley wire wear measuring apparatus using image processing according to claim 1 of the present invention, the following effects can be obtained.
(i) Since it is a non-contact method, high-speed operation is possible, and a long distance section can be measured in a short period.
(ii) Since the line sensor is installed at a position away from the existing structure such as point, air section, anchor, etc., the existing structure is compared with the method of directly measuring the thickness of the trolley wire using a rotating roller and an optical sensor. It is not necessary to consider the collision with the sensor, and the measurement can be continuously performed even in the place where the existing structure exists.

(iii) It can be measured even in daylight.
(iv) Basically, it is possible to capture line sensor images in all sections, and it is possible to acquire trolley lines in the measurement section and image data of existing structures in the vicinity.
(v) There is no need to use special lighting.

(vi) Compared with the method using laser light, it is not necessary to consider the influence on the human body, and it is easy to handle.
(vii) Since it is not necessary to receive the reflected light of the trolley wire wear part by regular reflection, there is no trouble of performing precise alignment between the light source and the light receiving device.
(viii) Since the line sensor image of the measurement section remains, it is possible to confirm the problem by looking at the image of the portion where there is a problem with trolley wire wear.

  According to the trolley wire wear measuring apparatus using image processing according to claim 2 of the present invention, in addition to the effects (i) to (viii) above, imaging by the deviation of the trolley wire by the discriminant analysis binarization method. A favorable threshold value can be determined regardless of the intensity of the luminance, and there is an effect that more accurate wear measurement can be performed.

(1) Basic concept An object of the present invention is to provide a measuring apparatus for measuring wear of a trolley wire by image processing.
In the present invention, for example, as shown in FIG. 1, a line sensor 1 installed so as to look up vertically on the roof of the inspection vehicle 4 is used as an image input means.

The line sensor 1 is installed such that the scanning line is perpendicular to the traveling direction of the inspection vehicle 4 and is horizontal, and the scanning line crosses the trolley line 3.
An illuminating device 2 is installed on the roof of the inspection vehicle 4 in the vicinity of the line sensor 1, and the illuminating device 2 illuminates a trolley wire 3 imaged by the line sensor 1.

As the illumination device 2, a device that is more powerful than ordinary white illumination is used, and specifically, the trolley wire worn portion can be imaged whiter than its side surface. This is because the worn portion of the trolley wire is a portion of the trolley wire that has been scraped by a pantograph, so that the trolley wire has a higher gloss than the portion that has not been worn, and is imaged white compared to the side surface by applying strong illumination.
The luminance signal of the scanning line obtained from the line sensor 1 is output to the measurement personal computer 5 installed inside the inspection vehicle 4.

The personal computer 5 for measurement arranges these luminance signals in time series, for example, creates a line sensor image (planar image) shown in FIG. 2 and stores it in the recording device 6 as an input image.
The measurement personal computer 5 performs image processing on the input line sensor image and obtains the width of the worn portion of the trolley wire according to the following procedures (A) to (D).

(A) Removal of the sky part by binarization processing When the trolley line 3 is imaged by applying strong illumination to the trolley line 3 in the daytime, the sky is reflected in the background, and as shown in FIG. In the sensor image, the trolley wire wear portion 3a is imaged whiter than the side surfaces 3b and 3c, whereas the trolley wire side surfaces 3b and 3c are imaged black due to rust, soot and the like.

Here, when the binarization processing is performed on the line sensor image of FIG. 2, as shown in FIG. 3, the sky A and the trolley wire wear portion 3a become a white region, and the trolley wire side surfaces 3b and 3c become a black region. .
However, it is not possible to determine whether the white area is the trolley wire wear portion 3a or the sky A if it remains as it is.

Therefore, the daytime light measurement process described below is performed.
First, as shown in FIG. 3, binarization processing is performed on the entire line sensor image.
Next, paying attention to the fact that the white area of the sky A is imaged much wider than the trolley wire wear part 3a, the trolley line thickness parameter input from the outside and the actual trolley line thickness parameter are taken into account. Is used to calculate the thickness of the trolley line on the screen, and the white area that appears beyond this thickness is inverted line by line.

As a result, as shown in FIG. 4, the area of the sky A is inverted from the white area to the black area, and only the trolley wire wear portion is extracted as the white area.
In FIG. 4, the trolley wire wear portion 3 a is not reversed and remains as a white region because the thickness of the trolley wire is as follows.

(A) Noise removal from the binarized line sensor image When a binarized line sensor image is constructed from the line sensor image by binarization processing, the trolley wire wear part is scratched and the dots are finely punctured. May be included.
Therefore, binarization processing expansion and contraction processing is performed to remove these noises (not shown).

(C) Edge detection of trolley wire wear part Next, the edges 3d and 3e on both sides of the trolley wire wear part 3a shown in the white area on the binarized line sensor image from which noise and existing structures have been removed are shown in FIG. Detect as shown in

For example, when searching for a certain scanning line from the left, the point where the background black changes to the white of the wear portion 3a is the edge point on the left side of the wear portion, and the point where the wear portion white changes to the background black is the wear portion. Detect as right edge point.
Then, by performing this process for each top-to-bottom line of the image, those edge points are continuous, and the edges 3d and 3e of the trolley wire wear portion 3a relating to one binarized line sensor image are detected. be able to.

(D) Calculation of trolley wire wear portion width Edges on both sides of one scanning line of the line sensor using edge data (see FIG. 5) on both sides of the trolley wire wear portion detected from the binarized line sensor image. The distance between points is calculated as the width of the trolley wire wear part on the image.
For example, the image resolution [mm / pix], which is the degree of the actual dimension [mm] for one image [pix], is calculated from the height from the line sensor 1 to the trolley line 3, the lens focal length, the sensor width, and the number of sensor pixels. Then, the width of the trolley wire wear portion 3a is calculated by multiplying the width of the trolley wire wear portion 3a on the image by the image resolution.

It is assumed that the height from the line sensor 1 to the trolley line 3 used for the calculation is obtained by another means. For example, the technology of the prior invention may be used.
The edge data thus obtained, the trolley wire wear part, the line sensor image used for the calculation, the data indicating the corresponding line number, etc. are recorded.

As described above, in the above-described embodiment, since it is a non-contact method, high-speed operation is possible, and a long distance section can be measured in a short time.
In addition, since the line sensor 1 is installed at a position away from the existing structure such as a point, an air section, and an anchor, the existing structure is compared with the method of directly measuring the thickness of the trolley wire using a rotating roller and an optical sensor. Therefore, it is possible to capture line sensor images in all sections, and to acquire image data of trolley lines in the measurement section and existing structures in the vicinity.

In addition, there is no need to use special illumination, and there is no difficulty in handling in consideration of the effects on the human body, compared to the case of using a sodium lamp or laser light, and there is no need for precision between the light source and the light receiving device. There is no hassle of performing proper alignment.
Furthermore, since daytime light measurement processing is performed, a trolley wire wear part can be extracted and measured not only at night but also in daytime.
Further, since the line sensor image remains as data, it is possible to confirm the problem of the portion having a problem as the trolley wire wear by looking at the image of the portion.

(2) Method in which discriminant analysis binarization method is added In the binarization process, when a threshold value is determined as a fixed value, depending on the environment at the time of imaging, other than the worn portion of the trolley line is emphasized / extracted, or the worn portion There is a problem that it cannot be extracted.
Here, the discriminant analysis binarization method is used to cope with the displacement of the trolley line and the difference in the intensity of the reflected light from the trolley line.

The discriminant analysis binarization method is a method of determining a threshold according to an image, and there is a “multiplier of mass multipliers gathered with a certain range of luminance values” (hereinafter referred to as a class) in a histogram in each image. The threshold value is determined so that the variance ratio between the intra-class variance and the inter-class variance, which is viewed in the background and the pattern area, is maximized (Non-Patent Document 2).
With this process, a relatively good threshold value can be determined for any image, and a wear part can be extracted.

FIG. 7 shows a trolley wire wear measuring apparatus using image processing according to the first embodiment of the present invention.
This embodiment relates to a specific configuration of the measurement personal computer 5 described in the basic concept.

  That is, the measurement personal computer 5 includes a line sensor image creation unit 10, a binarization processing unit 20, an empty region removal processing unit 30, a noise removal processing unit 40, and a trolley wire wear unit edge detection unit 50, as shown in FIG. The trolley wire wear part width calculation unit 60 and the like are used to determine the width of the wear part of the trolley wire according to the flowchart shown in FIG.

  First, the line sensor image creating unit 10 obtains a line sensor image by arranging the luminance signals of the scanning lines obtained from the line sensor 1 in time series (step S1). The created line sensor image is stored in the recording device 6 via the memory 70 and is sent to the binarization processing unit 20 via the memory 80 together with the parameters.

Next, the binarization processing unit 20 performs binarization processing on the entire line sensor image to obtain a binarized line sensor image (step S2). When the trolley line is imaged in the daytime, as shown in FIG. 3, in the binarized line sensor image, the white area is the empty A part or the trolley line wear part 3a, and the black area is the trolley line. Side surfaces 3b and 3c.
The binarized line sensor image is sent to the empty area removal processing unit 30 through the memory 80 together with the parameters.

The empty area removal processing unit 30 calculates the thickness of the trolley line on the screen using the trolley line height and the actual trolley line thickness input from the outside as parameters, and the white area that is equal to or larger than this thickness. The area is inverted for each line (step S3).
Since the sky A is equal to or larger than the thickness of the trolley line, it is inverted into the black area by the daytime light measurement process. Therefore, only the trolley wire wear part 3a is extracted as a white region.
The binarized line sensor image subjected to the daytime light measurement process is sent to the noise removal processing unit 40 via the memory 80.

The noise removal processing unit 40 removes fine dot-like noise included due to scratches on the trolley wire wear portion and the state of the background portion by expansion / contraction processing of binarization processing (step S4).
The binarized line sensor image subjected to the noise removal process and the daytime light measurement process is sent to the trolley wire wear part edge detection part 50 through the memory 80 together with the parameters.

  When the trolley wire wear part edge detection unit 50 searches for a certain line from the left, the point where the background black changes from the black of the wear part to the white of the wear part is the edge point on the left side of the wear part, and from the white of the wear part to the background black The process of detecting the changing point as the edge point on the right side of the worn part is performed for each line from the top to the bottom of the image, thereby detecting the edge of the trolley line worn part related to the binarized line sensor image (step S5).

Edges on both sides of the detected trolley wire wear portion are sent as edge data to the trolley wire wear portion width calculation unit 60 together with parameters via the memory 80.
The trolley wire wear portion width calculation unit 60 uses the edge data on both sides of the trolley wire wear portion detected from the binarized line sensor image to calculate the distance between the edge points on one scan line of the line sensor. Calculated as the width of the line wear portion on the image (step S5).

FIG. 9 shows a trolley wire wear measuring apparatus using image processing according to the second embodiment of the present invention.
This embodiment relates to a specific configuration of the measurement personal computer 5 described in the method in which the discriminant analysis binarization method is added.

Specifically, a discriminant analysis binarization processing unit 21 is provided instead of the binarization processing unit 20 in FIG. 7, and the line sensor image is image-processed according to the flowchart shown in FIG. Is what you want.
Here, the discriminant analysis binarization processing unit 21 performs discriminant analysis binarization processing in which a threshold is determined according to an image, that is, intra-class variance and inter-class variance for the background and pattern area when binarized. After determining the threshold value so that the variance ratio becomes maximum, the entire line sensor image is binarized to obtain a binarized line sensor image (step T2).

By this processing, a relatively good threshold value can be determined for any image, and the worn portion of the trolley wire can be extracted.
The other configurations of the present embodiment are the same as those of the apparatus configuration shown in FIG. 6, and each process is performed according to the flowchart shown in FIG.

  The line sensor image creation unit 10, the binarization processing unit 20, the discriminant analysis binarization processing unit 21, the empty region removal processing unit 30, the noise removal processing unit 40, the trolley wire wear unit edge detection unit 50, and the trolley wire wear. The copy width calculation unit 60 and the like may be realized by hardware or may be realized by software.

  The present invention is used in the field of measuring wear of a trolley wire by image processing, particularly measuring the width of a worn portion of the trolley wire.

It is a side view which shows the example of installation of the line sensor for trolley wire wear measurement. It is explanatory drawing which shows the line sensor image which imaged the trolley wire wear part in the daytime. It is explanatory drawing which shows what binarized the line sensor image shown in FIG. It is explanatory drawing which shows what performed the sky inversion process with respect to the line sensor image shown in FIG. It is explanatory drawing which shows what detected the edge of the trolley wire wear part from the line sensor image shown in FIG. It is a flowchart of the trolley wire wear measurement by a basic concept. It is a block diagram of a trolley wire wear measuring device based on a basic concept. It is a flowchart of the trolley wire wear measurement by the method which added the discriminant analysis binarization method. It is a block diagram of the trolley wire wear measuring apparatus by the method which added the discriminant analysis binarization method.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Line sensor 2 Illuminating device 3 Trolley line 4 Inspection vehicle 10 Line sensor image preparation part 20 Binarization process part 21 Discriminant analysis binarization process part 30 Empty area removal process part 40 Noise removal process part 50 Trolley line wear part edge detection Part 60 Trolley wire wear part width calculation part

Claims (2)

  Means for creating a line sensor image using a line sensor and storing it as an input image; means for performing binarization processing on the entire line sensor image; and for the line sensor image subjected to binarization processing Calculate the thickness of the trolley line on the screen using the trolley line height parameter input from the outside and the actual trolley line thickness parameter, and invert the larger white area to the black area Means for performing reversal processing, means for performing noise removal processing on the line sensor image subjected to binarization processing and reversal processing, and the line sensor image subjected to binarization processing, reversal processing and noise removal processing The white region is a trolley wire wear portion, and includes means for detecting an edge of the trolley wire wear portion, and means for calculating the width of the trolley wire wear portion from the detected edge. Trolley wire wear measuring device by image processing to.   Means for creating a line sensor image using a line sensor and storing it as an input image; means for performing discriminant analysis binarization processing on the entire line sensor image; and the line sensor subjected to discriminant analysis binarization processing Calculate the thickness of the trolley line on the screen using the trolley line height parameter input from the outside and the actual trolley line thickness parameter for the image. Means for performing inversion processing to invert the region, discriminant analysis binarization processing, means for performing noise removal processing on the line sensor image subjected to inversion processing, discriminant analysis binarization processing, inversion processing, and noise removal A white area in the processed line sensor image is defined as a trolley wire wear portion, a means for detecting an edge of the trolley wire wear portion, and a width of the trolley wire wear portion from the detected edge Trolley wire wear measuring device characterized by comprising calculating a means for performing.