JP5380000B2 - Trolley wire wear amount measuring method and measuring system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method and a system for measuring the wear amount of a trolley wire capable of measuring the wear amount with high accuracy without any handwork while reducing the data processing load for calculating the wear amount. <P>SOLUTION: The average value of the background level on the outer side of a trolley wire sliding surface is calculated on the sliding surface data of one line corresponding to the signal waveform obtained based on the light reception signal, the first reference level higher than the average value and lower than the minimum level of the light reception signal is obtained with the average value being a target, the data of one screen is binarized at the first reference level, and the coordinates of isolated points on the screen corresponding to the noise are extracted. By removing the isolated points from the data for one screen, the sliding surface data for each line with the isolated point noise being removed therefrom is obtained. The width of the sliding surface corresponding to two parameters can be easily calculated to the sliding surface data of one line. <P>COPYRIGHT: (C)2010,JPO&amp;INPIT

Description

  The present invention relates to a trolley wire wear amount measuring method and measurement system, and more specifically, the wear amount of a trolley wire uneven wear portion occurring at a branch point (branch point) of a railroad track is high accuracy and wear without manual operation. The present invention relates to a trolley wire wear amount measuring method and a measurement system capable of performing measurement while reducing the data processing load of amount calculation.

There are main lines and branch lines, ascending lines and descending lines, etc. as branch points (branch points) of the railway line. Between these trolley lines, the main line (upbound line) goes to the branch line (downbound line). Or, conversely, the vehicle moves from the branch line (down line) to the main line (up line) via the interference line, the pantograph of the vehicle comes into contact, and the trolley lines are taken over from each other.
Since the wire is not a power supply line to the vehicle, it is inclined upward and is pulled by a counterpoise (balancer) provided on a utility pole or the like. In order to prevent such an interference line from descending to a height similar to that of a pantograph or dropping below it, an unwanted contact accident with the pantograph and damage to the pantograph have been conventionally performed. A line detector is attached to the horn of the pantograph of the inspection vehicle.

Although there are various types of interfering lines, FIG. 9 is a schematic explanatory diagram of the interfering lines. Between the main line (up line) trolley line and the branch line (down line) trolley line, FIG. This is an example of a wavy line that is bridged diagonally.
Reference numeral 31 denotes a main trolley line, and reference numeral 32 denotes a branch trolley line. Tracks 33 and 34 are provided below the main trolley line 31 and the branch trolley line 32, respectively. Reference numeral 35 denotes a crossing wire, and a track 36 is provided below the wire. The wire 35 is installed obliquely between the main line and the branch line and at a position higher than the main line trolley line 31 and the branch line trolley line 32.
For example, the vehicle runs from the right to the left in the drawing and the pantograph of the vehicle crosses the trolley line from the main line to the branch line, or conversely, the vehicle runs from the left to the right in the drawing and moves from the branch line to the main line. The pantograph crosses the trolley line.
As the rail line at the branch point of the railway line is thus higher than the main line (up line) and branch line (down line) trolley line, when the vehicle pantograph crosses the line, the pantograph horn When the part hits the trolley wire at the destination, the sliding surface of the trolley wire is inclined and worn, that is, uneven wear occurs.
In addition, the structure of the pantograph of the inspection vehicle for the wire is known (Patent Document 1), and the measuring method and measuring device for the trolley wire wear amount are also known (Patent Document 2).
JP 2003-327019 A JP 2005-271682 A

The method and apparatus for measuring the amount of trolley wire wear as disclosed in Patent Document 2 has a problem in that it is not possible to accurately measure the uneven wear of the trolley wire at the tangle. Therefore, the present situation is that the partial wear of the wire part is currently performed manually.
The reason is that the reflected light corresponding to the width of the sliding surface is accurate because the trolley wire at the crossing line is composed of a normal sliding surface with flat wear followed by an inclined uneven wear sliding surface. It is because it cannot be obtained. In other words, the reflected light of the laser beam of the trolley wire wear measuring device does not return accurately from the inclined sliding surface of uneven wear.
FIG. 8 is an explanatory view of uneven wear of the trolley wire generated in the wire portion and the received light signal.
The sliding surface of the trolley wire 20 with uneven wear generated in the wire portion shown in FIG. 8A is composed of a flat normal sliding surface 20a and a subsequent uneven wear sliding surface 20b. The sliding surface reflection signal R (detection signal) obtained from the received light signal corresponding to the reflected light is composed of the reflected signal 21a of the normal sliding surface 20a and the reflected signal of the uneven wear sliding surface that follows this. However, as shown in FIG. 8 (b), the reflected signal 21a of the normal sliding surface 20a and the reflected signal 21b of a part of the uneven wear sliding surface 20b that follows this are continuous, and the uneven wear sliding surface. A part of the detection signal 20b is isolated and becomes a reflected signal 21c, which causes a problem that the reflected signal 21c cannot be distinguished from the noise signal 21d.
In addition, what is calculated as the amount of trolley wire wear in the wire portion is the remaining diameter A corresponding to the normal sliding surface 20a and the remaining diameter B corresponding to the uneven wear sliding surface 20b. The original amount of trolley wire wear is usually the distance from the center of the trolley wire to the sliding surface, but currently the remaining diameter is also treated as the amount of wear, so the remaining diameters A and B here are the trolley wire. The amount of uneven wear will be described below.

In FIG. 8B, a part of the detection signal of the uneven wear sliding surface 20b does not have a signal waveform in which the sliding surface reflection signal R is continuously independent as the reflected signal 21c, but the normal sliding surface The signal is divided into a waveform region C of a signal, a signal subsequent thereto, and a waveform region D of a signal isolated therefrom, thereby forming two waveform regions.
In such a case, it is impossible to distinguish whether the waveform region D is a noise waveform or a partial waveform of uneven wear. Therefore, the width of the sliding surface and the width of the uneven wear sliding surface or the normal sliding surface are usually used. + There is a problem that the sum width of the uneven wear sliding surfaces cannot be accurately detected from the sliding surface reflection signal R.
Therefore, the amount of wear at the uneven wear location is currently measured and managed by hand measurement. However, there is a problem in that the measurement value by hand measurement of the uneven wear portion has individual differences and is troublesome.
The object of the present invention is to solve such problems of the prior art, and calculate the amount of wear with high accuracy without manual operation of the amount of wear of the trolley wire uneven wear portion at the branch point of the railway track. It is an object of the present invention to provide a trolley wire wear amount measuring method and measuring system that can measure the data processing load of the trolley wire.

The feature of the first invention to achieve such an object is that the width of the normal sliding surface, the width of the uneven wear sliding surface and the normal sliding detected based on the light receiving signal at the uneven wear portion of the trolley wire. A table storing the remaining diameter of the trolley wire as the amount of uneven wear with two parameters selected from the sum of the widths of the surface and the uneven wear sliding surface as parameters, and kilometer information on the uneven wear location,
The signal waveform of the received light signal obtained by setting the gate in the range where the reflected light from the sliding surface of the trolley wire and the reflected light from the background can be obtained with respect to the received light signal or an amplified signal is converted into a digital value. Obtained as one-line sliding surface data of a predetermined width and sequentially collected and recorded corresponding to the distance traveled by the inspection vehicle,
The sliding surface data collected based on the kilometer information is retrieved, and a plurality of one-line sliding surface data including the position of the uneven wear point are read out, and a plurality of one-line sliding surface data are obtained for each horizontal line. Allocate the mileage recorded with the sliding surface data of one line assigned to the line in the vertical direction to obtain the data for one screen corresponding to the signal level of the received light signal,
The average value of the light reception signal level of the reflected light from the background or a part of the reflected light in one line of sliding surface data of the data for one screen is calculated, and the sliding surface data in one line Detects the lowest level over zero level in the waveform area that continues from the normal sliding surface,
When the lowest level is larger than the average value, the level between the lowest level and the average value is set as the first reference level, and the sliding surface data for each line of the data for one screen is binarized to obtain two values for one screen. Get the valuation data,
In the screen data formed by the binarized data, the coordinates of the isolated point corresponding to the noise are extracted, and the isolated point noise is removed from the data for one screen based on the coordinates of the isolated point.
Obtain one line of sliding surface data from which isolated point noise has been removed from the data for one screen, calculate the width of each sliding surface corresponding to two parameters from this sliding surface data, and refer to the table Thus, the remaining diameter of the portion corresponding to the sliding surface data of one line is obtained.

  The second invention is applied when the lowest level is smaller than the average value in the first invention, and further detects the highest level in the detection of the lowest level for the sliding surface data of one line, and the lowest level. The level between the highest level and the highest level is set as the second reference level, and the second reference level is set for each line of sliding surface data to obtain binarized data for one screen. The image area of the uneven wear location is searched from the data for one screen, and a straight line connecting the two points of the uneven wear start location and the uneven wear end location is set in the uneven wear location image region, and the horizontal line intersecting this straight line is set. With respect to the sliding surface data for one line, a straight line is used as the end of the uneven wear sliding surface, the sliding surface width corresponding to the two parameters is calculated, and the table is referenced to obtain the sliding surface data for this one line. Correspondence It is intended to obtain the remaining size of that location.

  Thus, in the first invention, the average value of the background level outside the trolley wire sliding surface is calculated in one line of sliding surface data corresponding to the signal waveform obtained based on the received light signal, Using this average value as a guide, a first reference level that is higher than this and lower than the lowest level of the received light signal is obtained, and the data for one screen is binarized at this first reference level to correspond to noise. Extract the coordinates of isolated points. Then, by removing the isolated points from the data for one screen, the sliding surface data for each line from which the isolated point noise has been removed is obtained. Then, the width of the sliding surface corresponding to the two parameters can be easily calculated with respect to the sliding surface data of one line. Therefore, by referring to the table with respect to the width of the two sliding surfaces, the remaining diameter at the location corresponding to the sliding surface data of one line can be calculated.

In addition, when the first reference level is lower than the average value of the background level, the wear amount may be obtained manually only at that portion as in the prior art, but by applying image processing by applying the second invention. The amount of uneven wear can be calculated.
In this case, the data for one screen is obtained by binarizing the sliding surface data for each line using the intermediate value between the highest level and the lowest level as the second reference value in the sliding surface data for one line. . Thereby, the binarized image showing the amount of uneven wear according to the reflection state of each one line is obtained. By counting the number of pixels from this image, the width of the normal sliding surface and the width of the uneven wear sliding surface or the sum of these widths can be easily obtained.
Therefore, by using this binarized image, the width of the sliding surface corresponding to the two parameters is calculated for each one line of sliding surface data, and this table is referenced to correspond to this one line of sliding surface data. It is possible to obtain the remaining diameter of the place to be.
As described above, in the first invention, the width of the normal sliding surface and the width of the uneven wear sliding surface can be easily obtained without performing special image processing. Since this process only needs to be performed when the first reference level is lower than the average value of the background level, the data processing load for wear amount calculation is reduced.
As a result, it is possible to realize a trolley wire wear amount measuring method and measurement system capable of easily measuring the wear amount of the trolley wire uneven wear portion at the branch point (branch point) of the railway track.

FIG. 1 is a block diagram of a measurement system for measuring the uneven wear amount of a trolley wire according to an embodiment to which the trolley wire wear amount measuring method according to the present invention is applied. FIG. 3A is a flowchart of the first half of the trolley wire uneven wear amount calculating process, FIG. 3B is a flowchart of the latter half of the trolley wire uneven wear amount calculating process, and FIG. 4A is an average value of the background level. FIG. 4B is an explanatory diagram of a dark level region for obtaining a background level, FIG. 5 is an explanatory diagram of isolated point noise and template processing for removing this, and FIG. FIG. 7 is a diagram for explaining the relationship between the moving surface data and the two slice levels, and FIG.
Reference numeral 1 denotes a trolley wire, which has a sliding surface 1a with which the pantograph contacts. Reference numeral 2 denotes a light receiving element composed of a one-dimensional CCD sensor or the like, which receives the reflected light from the trolley wire 1. Reference numeral 3 denotes an optical scanning projection system that irradiates the laser beam 3a perpendicularly to the sliding surface 1a over the displacement range of the trolley wire 1 through a mirror. The light receiving element 2 receives the reflected light from the trolley wire 1.
The laser beam 3a repeatedly scans the trolley line 1 in a direction crossing the rail, for example, at a frequency of 1,500 Hz, and scans the entire displacement range of the trolley line 1.

The light receiving element 2 is arranged in a direction in which its longitudinal direction crosses the rail, and its light receiving width has a light receiving range that can cover a range in which the trolley wire 1 is displaced in a zigzag manner.
Note that the pixel position of the light receiving element 2 on the left side of the drawing corresponds to the scanning start position of the optical scanning projection system 3, and the pixel position on the right side corresponds to the scanning end position of the optical scanning projection system 3. .
The optical scanning light projecting system 3 starts scanning with the laser beam 3a from the scanning start position in synchronization with the start signal ST. The start signal ST is output from the sliding surface data recording system processing device 10 and is also input to the data format circuit 6.
Light receiving element 2, optical scanning light projecting system 3, amplifier (AMP) 4, A / D conversion circuit (A / D) 5, data format circuit 6, distance pulse generation circuit 7, arithmetic unit 8, and sliding surface data The recording system processing device 10 is mounted on the inspection vehicle. Here, the sliding surface data of the trolley wire 1 is detected according to the traveling of the inspection vehicle, and is recorded corresponding to one scanning line.

The amplifier (AMP) 4 receives the light reception signal from the light receiving element 2, amplifies it, and inputs it to the A / D 5. The A / D 5 receives the light reception signal of the amplifier 4, digitizes the light reception signal (detection signal), and inputs it to the data format circuit 6. The distance pulse generation circuit 7 generates a distance pulse of one pulse for every 5 mm of travel distance, for example, according to the rotation of the wheels of the inspection vehicle.
The arithmetic unit 8 includes a DSP or MPU and a processing program, and includes a kilometer arithmetic processing block 8a and a gate pulse generation arithmetic processing block 8b.
The kilometer calculation processing block 8 a receives the distance pulse from the distance pulse generation circuit 7, calculates the travel distance as a kilometer, and outputs data corresponding to the current travel distance (kilometres) to the data format circuit 6. .
On the other hand, the gate pulse generation calculation processing block 8b generates a signal cut-out gate signal G (hereinafter referred to as gate signal G, high level ("H") significant) shown in FIG.
The gate signal G receives the output of the amplifier 4 and starts counting from the pixel signal at the start position, and detects the first light receiving pixel position where the reflected light is obtained in the light receiving element 2. The signal is “H” during this period in order to cut out a signal corresponding to the light receiving pixels from the pixel position 10 mm before the distance to the rear of the distance of 32 mm.
This gate signal G is supplied to the enable terminal e of the A / D 5 and the data format circuit 6. FIG. 2B is an explanatory diagram of the relationship between the gate signal G at this time and the sliding surface reflection signal R output from the amplifier 4 and D / A converted.
As shown in FIG. 2B, the sliding surface reflection signal R from the amplifier 4 is A / D converted by the A / D 5 and input to the data format circuit 6 while the gate signal G is “H”. Is done.

The data format circuit 6 has a data buffer memory 6a inside. The kilobyte data and the sliding surface reflection signal R are stored as digital values in the data buffer memory 6a in the format of bit data shown in FIG.
The format format is that the first 4 bits are about a kilobit record bit, the next 4 bits are the type code, and the sliding surface reflection from the trolley wire taken during the gate “H” period (see FIG. 2B). The signal R is sampled by A / D5 and converted to 8 bits to convert the multi-tone sliding surface reflection data Ri (however, the number of i-trolley line scans. Here, for convenience of explanation, one screen as described later. Is stored as).
The sliding surface reflection data Ri corresponds to one scan (hereinafter referred to as one line) of the laser beam 3a. When the A / D 5 is converted into 8 bits, 8 bits (1 byte) × 1600 data (1600 bytes) ) Is continuously stored. After that, 2 bits are provided as delimiter bits, and finally 4 bits are provided as a record of the next kilometer.
Here, the conversion amount of the sliding surface reflection data Ri for one line is set to 1600 in order to collect data for one pixel as 0.02 mm at 32 mm / 1600. When one screen is 1600 × 1200 pixels, 1600 pixels corresponds to the number of pixels for one horizontal line. On the other hand, 1200 samples corresponding to the vertical direction are collected corresponding to the travel distance (about kilometer) of the inspection vehicle. Here, the sliding surface reflection data Ri is i = 1 to 1200 and returns to i = 1 after 1200, and the data for the next screen corresponds to the travel distance (about kilometer) of the inspection vehicle. It shall be scanned and recorded.

  The data format circuit 6 receives the start signal ST from the sliding surface data recording system processing device 10 and synchronizes with the scanning of the light scanning type light projecting system 3 from the kilometer calculation processing block 8a to provide the current kilometer data 4 bits. Is stored in the first position of the data buffer memory 6a. When there are two trolley lines, the type code indicating which trolley line signal is used, and the next 4 bits are generated by recording, and the conversion data of A / D5 in accordance with the rise of the gate signal G The 8 bits are sequentially received, and 1600 pixels (the number of pixels for one horizontal line) of sliding surface reflection data Ri are sequentially stored in the next recording area of the data buffer memory 6a as digital values. Finally, the current kilometer data 4 bits are received from the kilometer calculation processing block 8a and stored in the last recording area of the data buffer memory 6a.

The A / D5 conversion clock CLK is synchronized with the start signal ST and has a sampling period exceeding 1600 in the period of the gate “H” with respect to the input data. The clock CLK generation circuit is omitted in the figure. Further, the data buffer memory 6a has an alternate use configuration in which when there are two areas and one area stores format data, the format data stored in the other area is read out.
Further, when there are two trolley lines, for example, the data format of FIG. 2A can be made to correspond to two trolley lines. In this case, for example, an area for recording the sliding surface reflection data Ri of one line of another trolley line may be provided by providing a type code before the delimiter bit is 2 bits.

The format data for the sliding surface reflection signal Ri of the data buffer memory 6a stored in the data formatting circuit 6 in this way is read to the sliding surface data recording system processing device 10 and the sliding surface data is stored in the area of the removable HDD 13. Will be stored continuously. In this case, i of the sliding surface reflection data Ri does not correspond to one screen, and a continuous trolley line scan count may be assigned.
The sliding surface data recording processing apparatus 10 includes an MPU 11, a memory 12, and a removable HDD 13, which are connected to each other via a bus 14. The memory 12 is provided with a format data writing program 12a for one line.
The MPU 11 executes the format data writing program 12a to read the format data including the sliding surface reflection data Ri from the area where the writing of the format data in the data buffer memory 6a is completed. A headline (for example, a line number, a data name, or a file name) such as a kilometer value is attached, and format data is sequentially written in a designated area of the removable HDD 13 corresponding to the kilometer value.
In this case, data for 1,200 lines corresponding to data for one screen may be recorded as one block by adding a heading such as a kilometer value at the head as one block.

Reference numeral 160 denotes a wear amount measuring system processing device, which includes an MPU 161, a memory 162, a screen memory 163, a CRT display 164, a keyboard 165, a removable HDD 13 of the sliding surface data recording system processing device 10, and the like. Are connected to each other.
The removable HDD 13 is one in which the above-described format data is recorded as the slick surface data in the slidable surface data recording system processing device 10, and here, this is inserted into the wear amount measuring system processing device 160.
When the removable HDD 13 is not removable, the sliding surface data recorded on the HDD 13 of the sliding surface data recording system processing device 10 is transferred from the sliding surface data recording system processing device 10 to the HDD 13 of the wear amount measuring system processing device 160. You can do it.
The memory 162 includes a tangential line location search program 162a, a background level average value calculation program 162b for calculating an average value of the background level outside the trolley line sliding surface, and a trolley line signal waveform. Minimum level / maximum level detection program 162c, slice level calculation program 162d, line correspondence binarization program 162e, noise correspondence isolated point removal program 162f, sliding surface width calculation program 162g, residual diameter calculation program 162h, binarization An alignment processing program 162i, a shift / smoothing processing program 162j, a partial wear location search program 162k, a straight line setting program 162l, a wear amount calculation program 162m of the partial wear location, etc. are stored, a remaining diameter calculation table 162n and a work area 162o, And Wataruri line position data table 162p are provided as required.

The trolley wire wear amount measurement process will be described below with reference to FIGS. 3A and 3B.
When the function key for the trolley wire wear amount measurement process is input from the keyboard 165, first, the interlaced wire location search program 162a is called and executed by the MPU 161.
The MPU 161 executes the interfering line location search program 162a. First, the kilometer of the interfering line location (for each line recorded by the sliding surface data recording system processing device 10 corresponding to the travel distance of the inspection vehicle). (Corresponding to the traveling distance of the inspection vehicle added to the sliding surface data) or the interference line position data of the interference line position data table 162p (corresponding to the traveling distance of the inspection vehicle as in the kilometer). It is obtained as a kilometer of the position of the wire to measure the uneven wear with reference (step 101).
Next, referring to the sliding surface recording data of the removable HDD 13 based on the kilometer of the interference line position data, the sliding surface reflection data Ri for one screen including the interference line is stored in the screen memory 163 (step 102). ). Then, the MPU 161 calls the background level average value calculation program 162b.
Next, when the background level average value calculation program 162b is executed by the MPU 161, the MPU 161 reads the sliding surface reflection data Ri for the first horizontal line from the screen memory 163 (step 103). Then, an average value level for 350 pixels is calculated for the region 17a for 350 horizontal pixels from the right end to 7 mm on one horizontal line of one screen shown in FIG. 4A (step 104).

4A and 4B are explanatory diagrams for calculating the background level average value, and 7 mm from the right end of each sliding surface reflection data Ri of FIG. 4A detected by the gate signal G stored in the screen memory 163. These 350 pixels will be described below with reference to FIG. 4B.
FIG. 4B shows the sliding surface reflection data Ri for one screen in which the sliding surface data portion is “H” and the other portions are low level “L” for the sake of simplicity, and the sliding surface is “H”. The area 15 of the surface reflection data Ri is set, and the front 10 mm is the “L” background level area 16, and the rear 7.3 mm is the “L” background level area 17. However, in practice, the sliding surface reflection data Ri is distorted on the sliding screen, so that the sliding surface reflection data Ri stored as each horizontal line in this one screen is the screen in the area 15 of the sliding surface. The upper position is slightly shifted left and right. Therefore, the area from which the background data can be acquired even with a slight shift is set here as the dark level acquisition area width from the last final pixel position (right end) of one horizontal line to 7 mm with a margin of 0.3 mm. .

Therefore, when the sliding surface reflection data Ri is read from the screen memory 163 for each horizontal line, the average of the levels for 350 pixels in the region 17a (see FIG. 4A) from the right end (final pixel position) to 7 mm from the screen. By calculating the value for each horizontal line, the average value Mi (where i is the horizontal line number) of the background level can be obtained for each sliding surface reflection data Ri.
The average value Mi is stored in the work area 162 o of the memory 162 for the sliding surface reflection data Ri of one horizontal line read in step 104. Then, the MPU 161 calls the trolley line signal waveform minimum level / maximum level detection program 162c.

When the minimum level / maximum level detection program 162c of the trolley line signal waveform is executed by the MPU 161, the MPU 161 reads the continuous signal waveform of the sliding surface next to the read sliding surface reflection data Ri of one horizontal line. The data of the lowest level Li and the highest level Hi is stored in the memory 162 in correspondence with the sliding surface reflection data Ri of one horizontal line detected by detecting the lowest level Li (where i is the horizontal line number) and the highest level Hi in the range. (Step 105). Then, the MPU 161 calls the slice level calculation program 162d.
Here, as shown in FIG. 8B, when the waveform area C and the isolated waveform area D are separated into two waveform areas, the waveform area C is obtained, and the lowest level Li in the waveform area C is obtained. And the highest level Hi.
When the two are not separated, the range of the continuous signal waveform on the sliding surface is the range of the region 15 of the sliding surface in FIG. 4B.
Note that the starting point of the continuous signal waveform range on the sliding surface is when it enters the level region above a certain value through the region 16 of the background level less than a predetermined value in the sliding surface reflection data Ri. is there. Then, the end of the range of the continuous signal waveform is from there until the level of the waveform signal is zero or substantially falls to zero.

I [Calculation of remaining diameter by normal processing] (processing from steps 101 to 123 shown in the flowcharts of FIGS. 3A to 3B)
After the step 105, when the slice level calculation program 162d is executed by the MPU 161, the MPU 161 compares the lowest level Li and the average value Mi of the background level to determine whether Li> Mi (step 106). ). If “YES” is determined here, the first slice level SAi (where i is the horizontal line number) is calculated as the slice level with respect to the sliding surface reflection data Ri of the horizontal line read out, and the background level is calculated by the following equation (1). It is set between the average value Mi and the lowest level Li.
SAi = Mi + (n / m) · (Li−Mi) (1)
However, n and m are arbitrary integers, and n <m, and are stored in advance in the parameter area of the memory 162 as parameters. Normally, n = 1, and m = 2, the sliding surface reflection data Ri for each horizontal line is calculated as an intermediate value between the background level average value Mi and the lowest level Li, and the read horizontal line sliding The first slice level SAi is stored in the memory 162 in correspondence with the surface reflection data Ri (step 107).

When the first slice level SAi of the read sliding surface reflection data Ri of the horizontal line 1 is stored in the memory 162, next, the horizontal line of the horizontal line read from the lowest level Li and the highest level Hi stored in the memory is stored. An intermediate value between them is calculated corresponding to the sliding surface reflection data Ri and calculated as a second slice level SBi (where i is a horizontal line number), and is stored in correspondence with the first slice level SAi. 162 (step 108). Then, the line-corresponding binarization program 162e is called.
The second slice level SBi here is a slice level for detecting the normal sliding surface width Y, and is determined as follows.
The second slice level SBi is calculated as SBi = Ni × K% + Li + α, where Ni is the difference between the highest level Hi and the lowest level Li of the sliding surface reflection data Ri. In this case, K is obtained by statistical processing to obtain a ratio of the signal levels of the reflected signal 21b and the reflected signal 21c in FIG. 8B to the maximum value, and + α ( It is the one that added extra). At this time, K is usually 10% to 25% with respect to Ni.
As shown in FIG. 4A, the sliding surface reflection data Ri has a large reflected light signal from the normal sliding surface first, and then the uneven wear sliding surface is inclined. The waveform gradually decreases at the boundary between the moving surface and the uneven wear sliding surface and moves in a small vertical direction according to the vertical movement of the inspection vehicle. Therefore, the second slice level SBi can be set as a level immediately above the maximum value of the uneven wear sliding surface by adding 10% to 25% to Ni from the lowest level Li.

Alternatively, the MPU 161 may obtain the second slice level SBi as SBi = SAi + OF obtained by adding a predetermined offset level OF to the first slice level SAi for each horizontal line.
The OF at this time is at a level exceeding the waveform area C (area including the reflected signal 21b and the reflected signal 21c in FIG. 8B) corresponding to the waveform uneven wear area in FIG. This is a constant value for selecting a level close to the maximum value of the signal level of the reflected signal 21b and the reflected signal 21c.
This can be selected as a constant value by obtaining the maximum value of the range in which the signal level of the reflected signal 21b and the reflected signal 21c varies by statistical processing, and using the value + α (allowance) obtained by this statistical processing as OF. It depends. Since the first slice level SAi varies corresponding to the sliding surface reflection data Ri at that time, the offset level OF can also be obtained as a value corresponding to this.

By the way, if it is determined in step 106 whether or not Li> Mi, and if NO, that is, if the lowest level Li of a certain horizontal line is Li <= Mi (except for Li = 0), The reflected light of the uneven wear amount is equal to or lower than the background level, so that almost no detection signal (sliding surface reflection signal) is obtained. The uneven wear amount of the trolley wire cannot be accurately obtained from the detection signal in such a case. Therefore, a flag Li <= Mi is set (step 106a), the number i of the horizontal line is stored in the memory 162 (step 106b), and the detection signal of the normal sliding surface width on the line and the normal sliding are detected. The detection signal of the deeeeeeeeeee width, which is the sum of the moving surface and the uneven wear sliding surface, is excluded, and the processing moves to the process of number i on the next horizontal line. Here, returning to step 103, the sliding surface reflection data Ri for the next horizontal line is read from the screen memory 163 and the same processing is performed.
When the flag is set in step 106a, II [calculation of remaining diameter by image processing] is executed by calling a binarization / alignment processing program 162i described later after I [calculation of remaining diameter by normal processing] is completed. Move on to processing. As a result, the width of the normal sliding surface and the sum of the normal sliding surface and the uneven wear sliding surface are obtained by image processing.

Next, the MPU 161 executes a line-corresponding binarization program 162e. When the line-corresponding binarization program 162e is executed by the MPU 161, the MPU 161 binarizes the read sliding surface reflection data Ri on the basis of the first slice level SAi, and 1 of the binarized image data. The line is stored in the work area 162o corresponding to the horizontal line number i (step 109). Then, it is determined whether or not the processing of the horizontal line for one screen has been completed (step 110). Here, since NO is initially determined, the process returns to step 103 and the same processing as described above is repeated. As a result, the binarization of each sliding surface reflection data Ri is repeatedly stored in the work area 162o of the memory 162 sequentially with each first slice level SAi as a reference.
When YES is determined in this step 110, the process of obtaining the binarized data for the sliding surface reflection data Ri (i = 1 to 1200) for one screen stored in the screen memory 163 is completed, and a part of the processing is obtained. In step 106a, the horizontal line number i is stored and some lines remain without being binarized. Normally, the data of one screen part is binarized at the first slice level SAi and stored as a binarized image in the work area 162o. Here, the MPU 161 calls the noise corresponding isolated point removal program 162f.

When the noise-corresponding isolated point removal program 162f is executed by the MPU 161, the MPU 161 performs pattern matching processing on the binarized image data stored in the work area 162o using the matrix mask 18 shown in FIG. 5B as a template. A process of extracting the coordinates (pixel position) of the isolated point on the image corresponding to the noise is performed (step 111).
As a result, the noise present in the background level regions 16 and 17 shown in FIG. 4B is removed from the sliding surface reflection data Ri shown in FIG. 4A, and the signal component in the region of the sliding surface 15 or the signal shown in FIG. The signal component of the waveform region C + D is left.
The region D of the signal waveform of the sliding surface with uneven wear that has become an isolated point shown in FIG. 8B is not matched with the matrix mask 18 of 7 pixels × 7 pixels, so that the coordinates are left without being extracted. . As a result, the reflected signal 21c of the uneven sliding surface 20b that cannot be distinguished from the noise isolated from the reflected signal 21a of the normal sliding surface 20a remains as a detection signal of the uneven wear sliding surface.

Here, the pattern matching process will be described using the matrix mask 18 as a template.
The matrix mask 18 has an outer 7 pixel × 7 pixel frame as an outer frame for white level detection, a 5 pixel × 5 pixel frame inside it is a middle frame for black level detection, and 3 pixels further inside it. This is a multi-frame template in which an inner frame of x3 pixels is a white level detection frame.
The image of the waveform region D corresponding to the reflected signal 21c in FIG. 8B is larger than the white point of the noise signal component in the binarized image shown in FIG. Therefore, the removal of this is performed by creating a matrix mask 18 of a matrix mask having a rectangular frame of 5 pixels × 5 pixels with a middle frame as shown in FIG. 5B, and moving the matrix mask 18 to each point on the screen. With this mask applied, all the pixels in the pixel area of the rectangular line of 5 pixels × 5 pixels in the middle frame are all white at a white level where only the pixels in the 3 pixels × 3 pixels area of the inner frame become “H”. The coordinates of the region of 3 pixels × 3 pixels are extracted when the pixel region has a pattern of a black level of L ″ and the outer frame of 7 × 7 pixels is not white. As a result, coordinate values of the white level surrounded by the black level, a maximum of nine coordinates are extracted as isolated point coordinates, and the coordinates of the isolated points for one screen are stored in the memory 162 corresponding to the line number of each horizontal line. This is a process of sequentially storing.

Next, with reference to the stored isolated point coordinates, the data of the isolated point coordinates in the multi-valued image of the screen memory 163 are all set to a black level value, and the noise component data is deleted.
Thereby, the reflected signal 21c of the uneven wear sliding surface 20b in FIG. 8B remains. Since the waveform area B is larger than the area of 3 pixels × 3 pixels in the inner frame, there is always a white level pixel that is “H” in the area of the rectangular line of 5 pixels × 5 pixels in the middle frame. It is. Alternatively, there is a 7 pixel × 7 pixel outer frame that is white.
Note that the sizes of the outer frame, the middle frame, and the inner frame of the matrix mask 18 are selected based on the size of the waveform region D generated in isolation, and the outer frame has 7 pixels × 7. The pixel is not limited to a pixel, and may not be a square such as 7 pixels × 5 pixels or 5 pixels × 7 pixels.

Therefore, next, the luminance level of the coordinate (pixel position) data extracted with respect to the sliding surface reflection data Ri of the screen memory 163 is set to zero, and this pixel is excluded from the sliding surface reflection data Ri (step 112). This eliminates isolated point noise from each sliding surface reflection data Ri of the screen memory 163.
In this noise-corresponding isolated point removal, assuming that the binarized data forms one image shown in FIG. 5A, a white point portion that is “H” becomes a noise signal component. In view of this, the coordinates of the white point are extracted and the level of the corresponding coordinate position of the sliding surface reflection data Ri in the screen memory 163 is changed to the black level and removed.
In the screen of FIG. 5A, the number of lines in the horizontal direction is different from the actual one for the sake of explanation, and is reduced.

When the execution of the noise corresponding isolated point removal program 162f by the MPU 161 is completed, the MPU 161 calls and executes the sliding surface width calculation program 162g. As a result, the MPU 161 reads the sliding surface reflection data Ri from which noise has been removed from the image memory 163 (step 113), and refers to the horizontal line number i stored in step 106b so that the lowest level Li is the average value Mi. The following determination is made (step 114), and if NO here, the sliding surface reflection data Ri is sliced at the first slice level SAi, and the first coordinate value X1 on the slice level in one horizontal line is the last. Is calculated (step 115), and then the sliding surface reflection data Ri is sliced at the second slice level SBi, and the first coordinate value Y1 and the last coordinate value on the slice level in one horizontal line are calculated. Y2 is calculated (step 116).
When the determination in step 114 is YES, the process returns to step 113 to read the sliding surface reflection data Ri corresponding to the next horizontal line and the same processing is repeated.

Here, as shown in FIGS. 6A and 6B, the first coordinate value X1 and the last coordinate value X2 on the slice level are calculated by slicing with the first slice level SAi. Even when the signal waveform of the sliding surface is not continuously separated as in (a), the signal waveform of the sliding surface is separated into two as shown in FIG. Even in the case of the present state, the total sliding surface width X (the sum of the normal sliding surface width Y and the uneven wear sliding surface width Z) can be calculated.
Therefore, next, the sum X of the normal sliding surface width Y and the uneven wear sliding surface width Z is defined as the total sliding surface width X,
Calculated by X = X2-X1,
Next, the normal sliding surface width Y is
Y = Y2−Y1 is calculated (step 117).
Then, the remaining diameter calculation program 162h is called.

Next, when the remaining diameter calculation program 162h is executed by the MPU 161, Z = X−Y is calculated, and the sliding surface reflection data Ri read based on whether or not the uneven wear sliding surface width Z is within the error range. It is determined whether or not is the data of the uneven wear portion (step 118).
In this determination, when Z <= T, the same applies to the determination of the horizontal line sliding surface reflection data Ri in the previous step 118. The total sliding surface width X is determined and the distance from the outer circumference corresponding to the sliding surface of the trolley wire to the sliding surface remains from the effective diameter before wear of the trolley wire and the total sliding surface width X. The wear amount is calculated by calculating the diameter or the distance from the center of the trolley wire to the sliding surface as the remaining radius (step 119).
Note that the effective diameter of the trolley wire before wear is known and stored in advance in a parameter area (not shown) of the memory 162.
Here, T is the maximum value measured as an error on a normal sliding surface in a state where uneven wear has not occurred.
If Z> T and two or more horizontal lines continue in the determination of step 118, then Z> T, YES is determined here, and it is determined that the part is unevenly worn. Therefore, the remaining diameter calculation table 162n is referred to (step 120), and as shown in FIG. 6A, the total sliding surface width X (normal sliding surface width Y and normal sliding surface width + uneven wear sliding). The remaining diameter stored at the location corresponding to the sum of the surface widths) is calculated as the wear amount, and corresponds to the normal sliding surface width Y and the total sliding surface width X of the sliding surface reflection data Ri for each horizontal line. And stored in the memory 162 (step 121).
Here, the remaining diameter calculation table 162n has a total sliding surface width X as the vertical axis column and a normal sliding surface width Y as the horizontal axis column. Is a data table recorded.

Next, it is determined whether or not the processing of the horizontal line for one screen has been completed (step 122). Since the first NO is determined here, the process returns to step 113, and the sliding surface reflection data Ri of the next horizontal line is It is read out and the same processing as described above is repeated. When the remaining diameter A and the remaining diameter B for one screen are calculated for each sliding surface reflection data Ri and the data processing for one screen stored in the screen memory 163 is completed, YES is returned to step 122. Here, the process proceeds to the flag determination process in step 123, where it is determined whether there is a flag set in step 106a.
Here, if the flag is not set and the answer is NO, the process returns to step 101, where the MPU 161 executes the interfering line location search program 162a, and inputs or interpolating about the kilometer of the interfering line location at the next location. With reference to the position data table 162p, the position of the tangent line for measuring the uneven wear is obtained as kilometer, and the same processing as described above is repeated until the position of the tangled line where measurement is required is completed.
On the other hand, if the flag is set in the previous step 106a and the answer to step 123 is YES, the binarization / alignment processing program 162i is called and II [calculation of remaining diameter by image processing] (step 124 shown in the flowchart of FIG. 3B). ˜134), the remaining diameter is calculated for the part that has been partially worn.

Next, when the binarization / alignment processing program 162i is executed by the MPU 161, the MPU 161 first resets the flag (step 124), and the sliding surface reflection data for one screen stored in the screen memory 163. Each Ri is sliced at the second slice level SBi and binarized (step 125), and the binarized image is stored in the screen memory 163 (step 126). This is each sliding surface data (each binarized sliding surface reflection data Ri) of one screen shown in FIG. As in FIG. 5A, the number of lines in the horizontal direction is different from the actual one for the sake of explanation, and is reduced.
Next, the justification / smoothing process program 162j is called and executed by the MPU 161. The MPU 161 executes this program and aligns the center position of the reflected light receiving portion along the center of the horizontal line with respect to the sliding surface data of each horizontal line of the binarized image in the screen memory 163. Processing is performed (step 127). This is the sliding surface data of one screen shown in FIG.

Next, the direction in which the vehicle crosses is obtained from the coordinate value Y2 and Y1 of the sliding surface reflection data Ri calculated in step 116 with reference to the interference line position data table 162p, and binarized in the direction in which the vehicle crosses. A process of bringing the sliding surface data of each line of the screen data to one side is performed. This is because the sliding of the other horizontal 1 line is made to coincide with the first coordinate of the sliding surface data of one horizontal line with respect to the sliding surface data of each horizontal one line of the screen memory 163. This is a process of shifting with the first coordinate of the moving surface data. Then, a smoothing process is performed on the contour line of the portion corresponding to the reflected light on the trolley line sliding surface of the binarized screen data (step 128). This is the sliding surface data of one screen shown in FIG.
In addition, in the wire position data table 162p, the direction in which the trolley wire is in contact with the pantograph in addition to the kilometer of the position of each wire is recorded corresponding to the kilometer of the position of each wire. Has been.
When the smoothing process is completed, the uneven wear location search program 162k is called and executed by the MPU 161.

  Next, when the sliding surface width calculation program 162k is executed by the MPU 161, the MPU 161 searches the uneven wear point P on one screen shown in FIG. Coordinates are obtained (step 130), a straight line L connecting the points Q1 and Q2 is calculated (step 131), and each horizontal line that has been binarized at the uneven wear point P where the straight line L is calculated (1) The first coordinate value X1 of the “H” portion of the sliding surface data) and the coordinate value X2 on the straight line L are calculated (step 132), and then each horizontal line (sliding surface data) at the uneven wear point P is calculated. ), The first coordinate value Y1 and the last coordinate value Y2 of the “H” portion are calculated (step 133).

Next, the MPU 161 calls and executes the sliding surface width calculation program 162g. Here, the sum X of the normal sliding surface width Y and the uneven wear sliding surface width Z for each horizontal line in the uneven wear portion P is defined as the total sliding surface width X,
Calculated by X = X2-X1,
Next, the normal sliding surface width Y is
Y = Y2−Y1 is calculated (step 134).
Then, the remaining diameter calculation program 162h is executed by the MPU 161, and the process returns to Step 120, where the remaining diameter calculation table 162n is referred to for each horizontal line at the uneven wear point P (Step 120), and each horizontal line at the uneven wear point P is referred to. The remaining diameter of the line is stored in the memory 126 and the process proceeds to step 122. Here, it is assumed that the processing of the uneven wear portion P for one screen has been completed, and the routine proceeds to step 123, where the flag is reset, so that the determination here is NO, and the routine returns to step 101. Then, the MPU 161 executes the interfering line location search program 162a, inputs the kilometer of the next interlaced wire location, or refers to the interpolated wire position data table 162p to find the kilometer of the interfering wire location. Then, the same processing as described above is repeated until the interfering wire portion requiring measurement is completed.
In the execution of the remaining diameter calculation program 162h by the MPU 161 here, I [normal processing is performed according to the normal sliding surface width Y and the total sliding surface width X of the sliding surface reflection data Ri of each horizontal one line. In addition to the calculated value of the calculation of the remaining diameter, the remaining diameters A and B are stored in the memory 162 as the uneven wear location P for each horizontal line at the uneven wear location P.

As described above, in the embodiment, the remaining sliding surface calculation table 162n is referred to by the total sliding surface width X of the normal sliding surface width Y, the normal sliding surface width Y, and the uneven wear sliding surface width Z. However, if the remaining diameter calculation table 162n is a search table for the uneven wear sliding surface width Z and the total sliding surface width X, the uneven wear sliding surface width Z and the total sliding are calculated. The remaining diameter can be calculated from the surface width X.
Furthermore, if the remaining diameter calculation table 162n is a search table for the normal sliding surface width Y and the uneven wear sliding surface width Z, the remaining diameter is determined by the normal sliding surface width Y and the uneven wear sliding surface width Z. Can be calculated.
Therefore, according to the present invention, the trolley wire has two parameters selected from the width of the normal sliding surface, the width of the uneven wear sliding surface, and the sum of the normal sliding surface and the uneven wear sliding surface as parameters. It is only necessary to calculate two parameter values in accordance with a table in which the remaining diameter of the item is stored as the amount of uneven wear.
9 is merely an example, and it is needless to say that the present invention can be applied to the measurement of the amount of trolley wire wear at various portions of the wire.

In the embodiment, two slice levels SAi and SBi are calculated, but this may be performed by inputting an optimum value from the CRT display 164.
Further, the second slice level SBi set between the lowest level and the highest level may be calculated as SBi = Hi × K% + Li + α based on only the highest level.
Furthermore, in the embodiment, the trolley wire uneven wear amount is calculated by I [calculation of remaining diameter by normal processing] or II [calculation of remaining diameter by image processing]. In this case, I [by the normal processing] Of course, only the calculation of the remaining diameter may be performed, and the amount of wear may be measured manually only for unknown ones.

FIG. 1 is a block diagram of a measurement system for measuring the uneven wear amount of a trolley wire according to an embodiment to which the trolley wire wear amount measuring method according to the present invention is applied. FIG. 2 is an explanatory diagram of the data that is formatted by the signal extraction gate signal and the data formatting circuit. FIG. 3A is a flowchart of the first half of a trolley wire uneven wear amount calculation process. FIG. 3B is a flowchart of the latter half of the trolley wire uneven wear amount calculation processing. FIG. 4A is an explanatory diagram for calculating the average value of the background level. FIG. 4B is an explanatory diagram of a dark level region for acquiring a background level. FIG. 5 is an explanatory diagram of isolated point noise and template processing for removing it. FIG. 6 is an explanatory diagram of the relationship between the sliding surface data for one line and the two slice levels. FIG. 7 is an explanatory diagram for calculating a partial wear amount by image processing. FIG. 8 is an explanatory view of uneven wear of the trolley wire generated in the wire portion and the received light signal. FIG. 9 is an explanatory diagram of the outline of the wire part.

Explanation of symbols

1 ... Trolley wire, 1a ... Trolley wire sliding surface,
2 ... light receiving element, 3 ... light scanning type light projecting system, 3a ... laser light,
4 ... Amplifier (AMP), 5 ... A / D conversion circuit (A / D),
6 ... Data format circuit, 6a ... Data buffer memory,
7 ... Distance pulse generation circuit, 8 ... Arithmetic unit,
8a: kilometer calculation processing block, 8b: gate pulse generation calculation processing block,
10: Sliding surface data recording system processing device,
11, 161 ... MPU, 12 ... memory,
12a ... Format data writing program,
13 ... Removable HDD, 14 ... Bus,
160 ... Abrasion amount measuring system processing device, ... MPU,
162 ... memory, 163 ... screen memory, 164 ... bus,
162a ... Interference line location search program,
162b ... Background level average value calculation program,
162c: trolley line signal waveform minimum / maximum level detection program,
162d: slice level calculation program,
162e ... a binarization program corresponding to the line,
162f: Noise corresponding isolated point removal program,
162 g ... sliding surface width calculation program,
162h ... residual diameter calculation program,
162i: binarization / alignment processing program,
162j: a program for shifting and smoothing,
162k: Uneven wear location search program,
162l ... straight line setting program, 162m ... wear amount calculation program for uneven wear location,
162n: Remaining diameter calculation table, 162o: Work area.

Claims (9)

In the trolley wire wear amount measuring method of irradiating light to the trolley wire, receiving reflected light from the sliding surface and measuring the wear amount of the trolley wire based on the received light signal,
One of the width of the normal sliding surface and the width of the uneven wear sliding surface detected based on the received light signal at the uneven wear portion of the trolley wire, and the width of the sum of the normal sliding surface and the uneven wear sliding surface A table in which the remaining diameter of the trolley wire is stored as an amount of uneven wear with two parameters selected from
The signal waveform of the light reception signal obtained by setting a gate in a range where the reflected light from the sliding surface of the trolley wire and the reflected light from the background can be obtained with respect to the light reception signal or an amplified signal is converted into a digital value. Converted and obtained as one line of sliding surface data of a predetermined width, sequentially collected and recorded according to the mileage of the inspection vehicle,
The sliding surface data collected based on the kilometer information is searched to read a plurality of one-line sliding surface data including the position of the uneven wear portion, and the plurality of one-line sliding Assigning surface data to each horizontal line, assigning the travel distance recorded together with the sliding surface data of the one line in the vertical direction, obtaining data for one screen in which the signal level of the received light signal corresponds to the brightness,
An average value of the signal level of the received light signal for the reflected light from the background or the reflected light of a part of the sliding surface data of the one line of the data for one screen is calculated, and the sliding of the one line is calculated. Detecting a minimum level of zero level or more in a waveform region continuous from the normal sliding surface in the moving surface data;
When the lowest level is larger than the average value, the level between the lowest level and the average value is set as a first reference level, and the sliding surface data of each one line of the data for one screen is binarized. To obtain the binarized data for the one screen,
In the screen data formed by the binarized data, the coordinates of the isolated point corresponding to the noise are extracted, and the isolated point noise is removed from the data for the one screen based on the coordinates of the isolated point,
The one-line sliding surface data from which the isolated point noise is removed from the data for one screen is obtained, and the width of each sliding surface corresponding to the two parameters is calculated from the sliding surface data. A trolley wire wear amount measuring method for obtaining the remaining diameter at a location corresponding to the sliding surface data of one line by referring to a table. In the detection of the lowest level, the highest level is detected in addition to the lowest level, the level between the lowest level and the highest level is set as a second reference level, and each sliding surface corresponding to the two parameters is detected. The width is calculated by calculating the difference between the first coordinate value and the last coordinate value in the horizontal line when the sliding surface data of the one line from which the isolated point noise has been removed is sliced at the first reference level. The horizontal line when the sliding surface data of the one line from which the isolated point noise has been removed is sliced at the second reference level is calculated as the sum of the normal sliding surface and the uneven wear sliding surface. the first coordinate values and the trolley wire wear measuring method of the last of the difference of coordinate values typically claim 1 and calculates the width of the sliding surface in.   The average value is obtained as an average value of luminance levels of a number of pixels in a region from a last pixel position on the horizontal line to a pixel position before a predetermined amount with respect to the sliding surface data of the one line. The trolley wire wear amount measuring method according to 1 or 2.   4. The trolley line according to claim 3, wherein the coordinates of the isolated points are extracted by searching the screen data formed of the binarized data with a matrix mask having a predetermined size over each point on the screen. Wear amount measurement method.   When the lowest level is smaller than the average value, the highest level is further detected in the detection of the lowest level for the sliding surface data of the one line, and the level between the lowest level and the highest level is set to the second level. The reference level is set, and the second reference level is set for each of the sliding surface data of one line to obtain binarized data for the one screen, and deviation from the binarized data for one screen is obtained. The image area of the wear part is searched, a straight line connecting the two points of the partial wear start part and the partial wear end part is set in the image area of the uneven wear part, and the slide of the one line for the horizontal line intersecting with the straight line is set. For the moving surface data, the sliding surface width corresponding to the two parameters is calculated using the straight line as the end of the uneven wear sliding surface, and the one-line sliding surface data is obtained by referring to the table. Trolley wire wear measuring method according to claim 1, wherein obtaining the remaining diameter of the point corresponding to the data.   The search of the image area of the uneven wear portion is performed by setting the center position of the reflected light receiving portion along the center of the horizontal line with respect to the sliding surface data of each one line of the binarized data for one screen. The binarized screen data is processed in a direction in which the vehicle crosses and the sliding surface data of each line of the binarized screen data is shifted to the trolley of the binarized data for one screen. 6. The trolley wire wear amount measuring method according to claim 5, wherein the trolley wire wear amount measuring method is performed on data for one screen after performing a smoothing process on a contour line of a portion corresponding to reflected light on a line sliding surface. In the trolley wire wear amount measurement system that irradiates light to the trolley wire, receives the reflected light from the sliding surface, and measures the wear amount of the trolley wire based on the received light signal,
The signal waveform of the light reception signal obtained by setting a gate in a range where the reflected light from the sliding surface of the trolley wire and the reflected light from the background can be obtained with respect to the light reception signal or an amplified signal is converted into a digital value. A trolley wire sliding surface data collection device that converts and obtains as one line sliding surface data of a predetermined width, and sequentially collects and records in accordance with the traveling distance of the inspection vehicle;
One of the width of the normal sliding surface and the width of the uneven wear sliding surface detected based on the received light signal at the uneven wear portion of the trolley wire, and the width of the sum of the normal sliding surface and the uneven wear sliding surface A trolley wire wear amount measuring device having a table storing the remaining diameter of the trolley wire as a partial wear amount with two parameters selected from
The trolley wire wear amount measuring device is:
The sliding surface data collected based on the kilometer information is searched to read a plurality of one-line sliding surface data including the position of the uneven wear portion, and the plurality of one-line sliding Assigning surface data to each horizontal line, assigning the travel distance recorded together with the sliding surface data of the one line in the vertical direction, obtaining data for one screen in which the signal level of the received light signal corresponds to the brightness,
An average value of the signal level of the received light signal for the reflected light from the background or the reflected light of a part of the sliding surface data of the one line of the data for one screen is calculated, and the sliding of the one line is calculated. Detecting a minimum level of zero level or more in a waveform region continuous from the normal sliding surface in the moving surface data;
When the lowest level is larger than the average value, the level between the lowest level and the average value is set as a first reference level, and the sliding surface data of each one line of the data for one screen is binarized. To obtain the binarized data for the one screen,
In the screen data formed by the binarized data, the coordinates of the isolated point corresponding to the noise are extracted, and the isolated point noise is removed from the data for the one screen based on the coordinates of the isolated point,
The one-line sliding surface data from which the isolated point noise is removed from the data for one screen is obtained, and the width of each sliding surface corresponding to the two parameters is calculated from the sliding surface data. A trolley wire wear amount measuring system that obtains the remaining diameter at a location corresponding to the sliding surface data of one line by referring to a table. In the detection of the lowest level, the highest level is detected in addition to the lowest level, the level between the lowest level and the highest level is set as a second reference level, and each sliding surface corresponding to the two parameters is detected. The width is calculated by calculating the difference between the first coordinate value and the last coordinate value in the horizontal line when the sliding surface data of the one line from which the isolated point noise has been removed is sliced at the first reference level. The horizontal line when the sliding surface data of the one line from which the isolated point noise has been removed is sliced at the second reference level is calculated as the sum of the normal sliding surface and the uneven wear sliding surface. trolley wire wear measuring system of the first difference of the coordinate values and the last coordinate values the normal claim 7 and calculates the width of the sliding surface in.   When the lowest level is smaller than the average value, the highest level is further detected in the detection of the lowest level for the sliding surface data of the one line, and the level between the lowest level and the highest level is set to the second level. The reference level is set, and the second reference level is set for each of the sliding surface data of one line to obtain binarized data for the one screen, and deviation from the binarized data for one screen is obtained. The image area of the wear part is searched, a straight line connecting the two points of the partial wear start part and the partial wear end part is set in the image area of the uneven wear part, and the slide of the one line for the horizontal line intersecting with the straight line is set. For the moving surface data, the sliding surface width corresponding to the two parameters is calculated using the straight line as the end of the uneven wear sliding surface, and the one-line sliding surface data is obtained by referring to the table. Trolley wire wear measuring system according to claim 7, wherein obtaining the remaining diameter of the point corresponding to the data.

Images