Summary of the invention
In above-mentioned classic method, there are some following problems.
Using the ruler such as vernier caliper to come in the method for direct measurement bay ceases to be busy thickness, the thickness of the expectation of pole line part is directly and reliably to measure.But, because the method does not automatically perform, so be difficult to measure the pole line of long section.
In the method for using the direct measurement bay ceases to be busy thickness of optical sensor,, measure so need carry out with low speed because measuring equipment contacts with pole line.In addition, because measuring equipment is clipped in the middle pole line, so when measuring equipment is hindered by existing structure (such as point, air section and anchor), can't use this measuring equipment.Therefore,, measuring equipment temporarily need be placed on place, so that prevent the collision between measuring equipment and the existing structure away from the measuring position in the place that has existing structure.
By coming with sodium vapor lamp or laser beam irradiation pole line in the method for measurement bay ceases to be busy wearing part width, it is necessary using specific illumination equipment such as sodium vapor lamp to be used for light source, and considers that specific illumination equipment also is necessary to the influence of human body.In addition, the method is easy to be influenced by noise, for example, described noise accompany the clip of pole line by the centre or the existing structure that comes across in the sensed image background caused.When under noise effect, obtaining erroneous measurements, can not write down the image of this moment, and therefore be not used in the method for checking this situation.Therefore, the part that causes erroneous measurements need be checked by the method for using direct measurement bay ceases to be busy thickness.The reflection of light light that receives the wearing part of irradiation pole line by direct reflection may be necessary, and therefore accurately positioned light source and printing opacity equipment are necessary.
Therefore, the purpose of this invention is to provide a kind of pole line wear measurement device, it is arranged to by Flame Image Process with the work efficiency that the improves width of measurement bay ceases to be busy wearing part correctly.
According to an aspect of the present invention, the pole line wear measurement device comprises: line sensor is suitable for striding pole line and scans; And the data processing unit that is connected in described line sensor, described data processing unit comprises: line sensor image creation parts are arranged to integrated data from described line sensor, to create the line sensor image; Image binarization processing element is arranged to the described line sensor image of binarization, to create the line sensor image of binarization; The noise removing processing element is arranged to from the line sensor image of described binarization and eliminates noise, has eliminated the line sensor image of the binarization of noise with establishment; Pole line wearing part rim detection parts are arranged to the edge of the pole line wearing part in the line sensor image that detects the described binarization of having eliminated noise; And pole line wearing part width calculating unit, be arranged to according to the width that calculates the pole line wearing part about the data of the height of the edge of detected pole line wearing part and pole line.Described pole line wearing part width calculating unit can be arranged to the execution following steps: select marginal point in the sweep trace in the described line sensor image as the reference marginal point; Determine the near linear of the marginal point in the preset range neighborhood of described reference edge point; Determine to comprise described reference edge point and second straight line vertical with described near linear; Determine the intersect edge point that another edge of described second straight line and the pole line wearing part relative with described reference edge point intersects; And the width that calculates the pole line wearing part according to the height of distance between described reference edge point and the described intersect edge point and pole line.Described pole line wearing part width calculating unit can be arranged to the execution following steps: first near linear of determining first group of marginal point in the preset range neighborhood of first marginal point in the sweep trace in described line sensor image; Determine second near linear of second group of marginal point in the preset range neighborhood of second marginal point in described sweep trace; Determine error degree with respect to first near linear of described first group of marginal point; Determine error degree with respect to second near linear of described second group of marginal point; Select a less conduct of the error degree that is associated in first marginal point and second marginal point with reference to marginal point; Determine to comprise described reference edge point and with the 3rd vertical straight line of the near linear that is associated; Determine the intersect edge point that another edge of described the 3rd straight line and the pole line wearing part relative with described reference edge point intersects; And the width that calculates the pole line wearing part according to the height of distance between described reference edge point and the described intersect edge point and pole line.Described pole line wear measurement device also can comprise: second line sensor, and the top that is suitable for striding bow collector (pantograph) is scanned; And second data processing unit that is connected in described second line sensor, described second data processing unit comprises: line sensor image creation parts are arranged to integrated data from described second line sensor to create the line sensor image; And bow collector top high computational parts, be arranged to the height that calculates pole line according to described line sensor image.
According to a further aspect in the invention, a kind of pole line abrasion inspection car comprises: the main body that wheel is arranged; The line sensor that is installed on described main body top and points to; And the data processing unit that is connected in described line sensor, described data processing unit comprises: line sensor image creation parts are arranged to integrated data from described line sensor, to create the line sensor image; Image binarization processing element is arranged to the described line sensor image of binarization, to create the line sensor image of binarization; The noise removing processing element is arranged to from the line sensor image of described binarization and eliminates noise, has eliminated the line sensor image of the binarization of noise with establishment; Pole line wearing part rim detection parts are arranged to the edge of the pole line wearing part in the line sensor image that detects the described binarization of having eliminated noise; And pole line wearing part width calculating unit, be arranged to according to the width that calculates the pole line wearing part about the data of the height of the edge of detected pole line wearing part and pole line.
According to a further aspect in the invention, a kind of pole line wear measuring method comprises: integrated data from described line sensor, to create the line sensor image; The described line sensor image of binarization is to create the line sensor image of binarization; From the line sensor image of described binarization, eliminate noise, eliminated the line sensor image of the binarization of noise with establishment; Detect the edge of the pole line wearing part in the line sensor image of the described binarization of having eliminated noise; And according to the width that calculates the pole line wearing part about the data of the height of the edge of detected pole line wearing part and pole line.
Embodiment
With reference now to Fig. 1, to Fig. 5, Figure 10 and Figure 11, shows pole line wear measurement device according to first embodiment of the invention.Fig. 1 is the perspective schematic view of the pole line wear measurement device of first embodiment.As shown in fig. 1, line sensor 11, line sensor 12, illuminating lamp 13 and bow collector 14 are installed in the top with inspection trolley 10 that wheel body is arranged.Illuminating lamp 13 is the common white light sources that are suitable for launching white light, is used for pole line 16 is thrown light on.First data processing unit 1 is connected in line sensor 11 and is used for signal communication.Second data processing unit 2 is connected in line sensor 12 and is used for signal communication.First data processing unit 1 and second data processing unit 2 interconnect.
Line sensor 11 is installed on the top of inspection trolley 10, and directed in orthogonal makes progress in the top plan of inspection trolley 10.Settle line sensor 11 in the following manner, promptly direction of scanning 18 is identical with the sleeper direction, and sweep trace strides across pole line 16.On the other hand, line sensor 12 is installed on the top of inspection trolley 10, and the top plan of sensing and inspection trolley 10 is tilted to.Settle line sensor 12 in the following manner, promptly extend along the vertical direction of inspection trolley 10 direction of scanning 19, and sweep trace strides across bow collector 14.
Fig. 5 is the exemplary system block diagram of the pole line wear measurement device of first embodiment.First data processing unit 1 comprises assembly 50 to 57.As shown in Figure 5, the line sensor picture signal F50 that line sensor 11 obtains for each sweep trace, concurrent line sending sensor image signal F50 is to line sensor image creation parts 50.Line sensor image creation parts 50 are arranged to time series settles line sensor picture signal F50 creating line sensor image F51, and is arranged to line sensor image F51 is stored in the storer 51.The line sensor image F51 that is stored in the storer 51 carries out relaying via data transmission link 57, and is stored in the storer 52.
Image binarization processing element 53 is arranged to being stored in line sensor image F51 application image binarization process in the storer 52 creating the line sensor image F52 of binarization, and the line sensor image F52 that is arranged to binarization is stored in the storer 52.
Noise removing processing element 54 is arranged to be eliminated process to the line sensor image F52 using noise that is stored in the binarization in the storer 52 and has eliminated the line sensor image F53 of the binarization of noise with establishment, and is arranged to the line sensor image F53 that will eliminate the binarization of noise and is stored in the storer 52.
Pole line wearing part rim detection parts 55 be arranged to be stored in elimination in the storer 52 the line sensor image F53 of binarization of noise use pole line wearing part edge detection process creating pole line wearing part marginal date F54, and be arranged to pole line wearing part marginal date F54 be stored in the storer 52.
Pole line wearing part width calculating unit 56 is arranged to uses pole line wearing part width computation process creating pole line wearing part width data F58 to being stored in pole line wearing part marginal date F54 in the storer 52 and pole line altitude information F55, and is arranged to pole line wearing part width data F58 is stored in the storer 52.Bow collector top altitude information F116 as described below is used to pole line altitude information F55.As shown in figure 11, being stored in bow collector top altitude information F116 in the storer 112 is transferred to storer 52 via data transmission link 57 and the data transmission link 118 that is connected in data transmission link 57 and is stored in the storer 52.
With reference to the process flow diagram of figure 4, the process of width of the wearing part of the pole line of measuring among first embodiment 16 has been described below.As shown in Figure 4, at first, at step S11, by the line sensor picture signal F50 of line sensor 11 acquisitions for each sweep trace (along the x axle), and it is placed in the time series (along the t axle), creating line sensor image F51 (two dimensional image) in the on-line sensor image creation parts 50, and line sensor image F51 is stored in the storer 51 as input picture.
After step S11, at step S12, be stored in line sensor image F50 in the storer 51, and it is stored in the storer 52 via data transmission link 57 relayings.Then, setting is used for the threshold value of binarization, and the image of the pole line wearing part of bar shaped is separated from one another with other background parts that comes across among the line sensor image F51 based on described threshold value by image binarization process, in image binarization processing element 53, to create the line sensor image F52 of binarization.Because the part of being worn and torn by bow collector 14 has constituted the wearing part of pole line 16, wearing part is not thicker so the wearing part of pole line 16 is than other.Therefore, among the on-line sensor image F51, the wearing part of pole line 16 appears as the bar shaped different with the brightness value of background parts.This Flame Image Process has been created the line sensor image F52 of binarization as shown in Figure 2.In the line sensor image 26 of binarization, the wearing part of pole line 16 is represented as white portion 20, and background parts is represented as black region 21.
After step S12, at step S13, noise removing processing element 54 is by carrying out expansion process and dwindling process and from by removing noise the line sensor image F52 that makes the binarization that line sensor image F51 binarization creates.This process is called as the noise removing process.This process is used as the spaced point that removes the noise that wearing part and background portion branch according to pole line 16 cause from the line sensor image F52 of binarization.
After step S13, at step S14, the left and right edges 22 of pole line wearing part rim detection parts 55 testing stand ceases to be busys 16, it is indicated as the white portion 20 among the line sensor image F53 of the binarization of having eliminated noise.This process is called as pole line wearing part edge detection process.When the sweep trace 23 in the line sensor image F53 of the binarization of having eliminated noise is searched for, change into white portion 20 parts of indication wearing part from the black region 21 of indication background parts in color, detect a point as left hand edge 24 (coboundary), and when the sweep trace 23 in the line sensor image F53 of the binarization of having eliminated noise is searched for, change into black region 21 parts of indicating background parts in color from the white portion 20 of indication wearing part, detect another point as right hand edge 25 (lower limb).Carry out this process fully for each sweep trace, eliminated the edge 30 of the wearing part of the pole line 16 among the line sensor image F53 of binarization of noise with detection.
After step S14, at step S15, according to the pole line wearing part marginal date F54 at the edge of the wearing part of the pole line of being created for line sensor image F53 16 based on the binarization of having eliminated noise, calculate for each sweep trace of line sensor 11, as the edge-to-edge distance 31 of the distance between the left and right edges on the image 30.This process is called as pole line wearing part width computation process.Calculate the image resolution ratio [mm/pix] of the actual size [mm] of each pixel [pix] that is defined as on the image based on the number of pixels of 16 height, the focal length of lens, sensor scan width and sensor from line sensor 11 to pole line.By being multiplied each other, the picture traverse of pole line wearing part and image resolution ratio calculate the width of pole line wearing part.On the record segment (not shown), write down the width of the marginal date of being calculated, the pole line wearing part that is calculated, the data of indicating the line sensor image that is used for the edge calculation data and the data that identify employed respective scan line.By bow collector top 15 being measured as 16 height calculates 16 the height F55 from line sensor 11 to pole line that is used for aforementioned calculation from sensor 11 to pole line.Transmit the height on bow collector tops 15 by data transmission link 57, and it is stored in the storer 52.
With reference to Figure 11, the scheme of the pole line wear measurement device of the height that is used to measure bow collector 14 has been described below.Line sensor 12 is installed on the front end of the bow collector 14 at inspection trolley 10 tops, so that measure the height of bow collector 14 in the following manner, described mode is that extend along the vertical direction of bow collector 14 direction of scanning.Second data processing unit 2 comprises parts 110 to 118.As shown in figure 11, line sensor 12 obtains the line sensor picture signal F110 for each sweep trace, and line sensor picture signal F110 is sent to line sensor image creation parts 110.Line sensor image creation parts 110 are arranged to time series settles line sensor picture signal F110 creating line sensor image F111, and is arranged to line sensor image F111 is stored in the storer 111.The line sensor image F111 that is stored in the storer 111 carries out relaying via data transmission link 118, and is stored in the storer 112.
Storer 112 is arranged to a series of line sensor images of storage and maintenance for a series of segments of pole line, and it is used to check following part, and described part is for being identified as problematic pole line wearing part after the operation of pole line wear measurement.
Image binarization processing element 113 is arranged to being stored in line sensor image F111 application image binarization process in the storer 112 creating the line sensor image F112 of binarization, and the line sensor image F112 that is arranged to binarization is stored in the storer 112.
Bow collector width filter processing element 114 is arranged to uses bow collector width filter process creating bow collector trace image F113 to the line sensor image F112 that is stored in the binarization in the storer 112, and is arranged to bow collector trace image F113 is stored in the storer 112.
Bow collector track following filtration treatment parts 115 are arranged to be used bow collector track following filter process and has carried out the bow collector trace image F114 of bow collector track following filtration treatment to create being stored in bow collector trace image F113 in the storer 112, and is arranged to the bow collector trace image F114 that will carry out bow collector track following filtration treatment and is stored in the storer 112.
Bow collector singularity (singularity) filtration treatment parts 116 are arranged to be used bow collector singularity filter process and has carried out the bow collector trace image F115 of bow collector singularity filtration treatment to create being stored in the bow collector trace image F114 that carries out bow collector track following filtration treatment in the storer 112, and is arranged to the bow collector trace image F115 that will carry out bow collector singularity filtration treatment and is stored in the storer 112.
Bow collector top high computational parts 117 are arranged to uses bow collector top high computational process creating bow collector top altitude information F116 to being stored in the bow collector trace image F115 that carries out bow collector singularity filtration treatment in the storer 112, and is arranged to bow collector top altitude information F116 is stored in the storer 112.
With reference to the process flow diagram of Figure 10, the process of measuring the top height of bow collector 14 in first embodiment has been described below.As shown in figure 10, at first, at step S21, by the line sensor picture signal F110 of line sensor 12 acquisitions for each sweep trace (along the y axle), and it is placed in the time series (along the t axle), creating line sensor image F111 (two dimensional image) in the on-line sensor image creation parts 110, and line sensor image F111 is stored in the storer 111 as input picture.
After step S21, at step S22, be stored in line sensor image F110 in the storer 111, and it is stored in the storer 112 via data transmission link 118 relayings.Then, setting is used for the threshold value of binarization, and the image of the bow collector 14 of bar shaped is separated from one another with other background parts that comes across among the line sensor image F111 based on described threshold value by image binarization process, in image binarization processing element 113, to create the line sensor image F112 of binarization.Because bow collector 14 is different with the thickness of background parts, so among the on-line sensor image F111, the image of bow collector 14 appears as the bar shaped part different with the brightness value of background parts.This Flame Image Process has been created the line sensor image F112 of binarization.In the line sensor image F112 of binarization, the image of bow collector 14 is represented as white portion, and background parts is represented as black region.
After step S22,, carry out bow collector width filter process at step S23.Particularly, check each sweep trace of the line sensor image F112 of binarization with the direction of scanning, to detect coboundary and lower limb.Because in the line sensor image F112 of binarization, the image of bow collector 14 appears as the bar shaped that has corresponding to certain constant width of the thickness of bow collector 14, so the white portion that has with the image same widths of bow collector 14 by detection identifies bow collector 14.When coboundary in a sweep trace and the interval between the lower limb in the width at preset range the time, it is detected and be designated bow collector 14.Lower limb is left white, indicating the upper end of bow collector 14, and other parts is changed into black.By repeating the track that said process identifies bow collector upper end 15.This process is called as bow collector width filter process.The image that shows the track of the bow collector apical position of creating in this process is called as bow collector trace image F113.But, following situation is possible: when the track by bow collector width filter process identifier bow collector top 15, in fact may not be also detected as bow collector apical position 15 corresponding to the part of bow collector apical position 15.This may take place in following situation, promptly, in the line sensor image F112 of binarization, the object except bow collector 14 (such as pole line 16, insulator and be used for the lead of hoisting frame ceases to be busy) appears as the image that has with bow collector 14 approximate width.In order to eliminate the part that remaining these error-detecting that become the noise among the bow collector trace image F113 arrive, carry out following steps.
After step S23,, carry out bow collector track following filter process at step S24.In the line sensor image F112 of binarization, bow collector 14 appears as the bar shaped with constant width.As a result, in bow collector trace image F113, bow collector 14 appears as the straight line of being made up of the series of points on bow collector top 15.In this step, the point that has than the shorter length of predetermined threshold is eliminated as noise.This process is called as bow collector track following filter process.But, bow collector track following filter process can be eliminated the noise of being made up of short point, but can't following noise section, and it has satisfied the condition of the length that is used to identify the bow collector apical position.Based on the following true following steps that comprise another noise removing process of carrying out, the described fact is: appear as one single on the bow collector 14 on-line sensor images, make on the sweep trace of on-line sensor 12, bow collector top 15 also appears as a single point, and in the bow collector trace image, the great majority point that is detected by bow collector width filter process all correctly indicates the point on bow collector top 15.
After step S24,, carry out bow collector singularity filter process at step S25.When detecting two or more tracing point in the sweep trace at a line sensor 12 of bow collector trace image, the most detected tracing point of the mean place of approaching serial tracing point is designated the real trace point, and eliminates detected other tracing point.This process is called as bow collector singularity filter process.
After step S25,, carry out bow collector top high computational process at step S26.Offset except the bow collector trace image of noise and change, to calculate the actual three-dimensional height on bow collector top.This conversion comprises the projection conversion based on the installation site of line sensor 12 and attitude.By connecting data transmission link 118 and data transmission link as shown in Figure 5 57 as shown in figure 11, in pole line wearing part width calculating unit 56, the bow collector top height of Que Dinging is used as pole line altitude information F55 as mentioned above.Be used for the situation safeguarding and check at measurement bay ceases to be busy height in advance, this measured value can be used as pole line altitude information F55.Particularly, storer 112 can be arranged to the altitude information that is stored in pole line wear measurement pole line 16 before.
With reference to figure 6 and Fig. 8, show pole line wear measurement device according to second embodiment of the invention.Fig. 6 is the exemplary system block diagram of the pole line wear measurement device of second embodiment.In pole line wear measurement device as shown in Figure 6, at the step S15 shown in the process flow diagram among first embodiment of Fig. 4,60 execution of pole line wearing part width calculating unit have the pole line wearing part width computation process to the compensation of the inclination of pole line.
In Fig. 8, comprise left hand edge point 80 as the sweep trace 87 of target.At first, the marginal point of left hand edge point 80 in the neighborhood 81 of preset range is approximately near linear 82.Comprise left hand edge point 80 and be defined as intersect edge point 85 perpendicular to the right hand edge point on the straight line 83 of near linear 82.The pole line wearing part width of on the left side edge point 80 is defined as the distance 86 between left hand edge point 80 and the intersect edge point 85.
With reference to figure 7 and Fig. 9, show pole line wear measurement device according to third embodiment of the invention.Fig. 7 is the exemplary system block diagram of the pole line wear measurement device of the 3rd embodiment.In pole line wear measurement device as shown in Figure 7, at the step S15 shown in the process flow diagram among first embodiment of Fig. 4,70 execution of pole line wearing part width calculating unit have another pole line wearing part width computation process to the compensation of the inclination of pole line.
In Fig. 9, comprise left hand edge point 90 and right hand edge point 91 as the sweep trace 102 of target.At first, left hand edge point 90 and right hand edge point 91 marginal point in the neighborhood 92 of preset range is approximately near linear 93 and 94 respectively.When the approximate error 95 of near linear 93 and less than the approximate error 96 of near linear 94 and in, it is to comprise left hand edge point 90 and perpendicular to the right hand edge point on the straight line 98 of near linear 93 that intersect edge point 99 is calculated as.Be defined as distance 100 between left hand edge point 90 and the intersect edge point 99 at the pole line wearing part width of sweep trace 102.On the other hand, when approximate error 96 and less than approximate error 95 and in, it is to comprise right hand edge point 91 and perpendicular to the left hand edge point on the straight line 101 of near linear 94 that the intersect edge point is calculated as.Be defined as distance between left hand edge point of crossing and the right hand edge point 91 at the pole line wearing part width of sweep trace 102.
Use the pole line wear measurement device of the Flame Image Process in the illustrated embodiment to be suitable for providing the pole line of electric power to carry out wear measurement to electric train or vehicle to being used for.
Advantage and effect that pole line wear measurement device by illustrated embodiment is produced have below been described.
(1) move owing to described pole line wear measurement device need not directly to contact with pole line, therefore described pole line wear measurement device is carried out wear measurement operation at a high speed, makes the pole line that can measure long distance at short notice.
(2) because described pole line wear measurement device is included in the locational sensor away from existing structure (such as point, air section and armature), therefore described pole line wear measurement device need not to consider the collision with existing equipment, make that described pole line wear measurement device is suitable for existing the place of existing structure for the method for using the next direct measurement bay ceases to be busy thickness of swing roller and optical sensor.
(3) because described pole line wear measurement device is created in the line sensor image in all sections of wanting, therefore can obtain the view data of pole line and the existing structure around described pole line neighborhood.
(4) described pole line wear measurement device need not to use the special lighting for light source, such as sodium vapor lamp.
(5) because described pole line wear measurement device is not used laser beam, therefore described pole line wear measurement device need not to consider the influence of laser beam to human body, makes for the method for using laser beam, is easier to operate described device.
(6) because described pole line wear measurement device does not receive the reflection of light light of the wearing part of irradiation pole line by mirror reflection, therefore described pole line wear measurement device need not accurately positioned light source and printing opacity equipment.
(7) owing to stored the view data in the measuring section, therefore can check the part that is identified as problematic pole line wearing part by using institute's memory image.
(8) produce following advantage except the advantage of first embodiment based on the described pole line wear measurement device of the Flame Image Process of second embodiment: when pole line tilts with respect to the moving direction of inspection trolley, compensate this inclination, so that determine the width of pole line wearing part more accurately.
(9) produce following advantage except the advantage of first embodiment based on the described pole line wear measurement device of the Flame Image Process of the 3rd embodiment: when pole line tilts with respect to the moving direction of inspection trolley, compensate this inclination, so that determine the width of pole line wearing part more accurately.Particularly, even when an edge of pole line wearing part is worn shape into wavy or a series of recesses, also can determine the width of pole line wearing part more accurately.
The application is based on the 2005-68796's of Japanese patent application formerly that is filed on March 11st, 2005.Be incorporated herein by reference at this full content this Japanese patent application 2005-68796.
Although below described the present invention with reference to some embodiment of the present invention, the present invention is not limited to embodiment described above.According to above instruction, those skilled in the art can realize the modifications and variations to embodiment described above.Define scope of the present invention with reference to following claim.