JP2004508643A - Method and apparatus for determining a change in a reference object - Google Patents

Abstract

The present invention relates to a method and an apparatus for determining a change in a reference object belonging to a travel path and provided at a predetermined location in the travel path. According to the invention, at least one reference image is recorded at a first point in time at a predetermined location on the road. An image of the reference object is formed in at least one of the reference images. At least one inspection image is recorded at a second point in time at a predetermined location on the travel path. Another image of the reference object is formed in at least one of the inspection images. By comparing the image of the reference object in the at least one reference image with another image in the at least one inspection image, a change of the reference object is determined.

Description

[0001]
The present invention relates to a method and an apparatus for determining a change in a reference object. A method or device of this kind is known from document [1]. Such an image processing method, that is, a method of encoding and decoding a digital image according to the block encoding standard MPEG4 [9] is based on object-based image encoding.
[0002]
In the case of object-based image coding, the segmentation of two temporally consecutive images is segmented into image blocks or image regions according to the object appearing in one scene, and the object is coded separately .
[0003]
In the case of this object-based image encoding, an image of an object is formed in so-called inter-image encoding that involves motion prediction in a temporally preceding image. A further image of the object is then formed in the subsequent image in time. Difference image information is determined using image information assigned to the image of the object in the temporally preceding image and image information assigned to another image of the object in the temporally subsequent image. Thereafter, the difference image information is encoded and transmitted.
[0004]
For example, if the change of the object in the digital image which is temporally successive due to the movement or shape change of the object, that is, the object, is small, the difference image information is negligible, and if the change is large, very large difference information is generated. . For this reason, the “movement for each image” (motion prediction) is measured and compensated before obtaining difference information (motion compensation), as known from document [1].
[0005]
Document [2] discloses an image processing method for monitoring a traveling route of a railway. According to this method, a video image of a trolley wire used for railway power supply is recorded. In this case, the trolley wire is guided along the travel path by using a longitudinal chain member, a so-called overhead wire, and a transverse support member, a bracket.
[0006]
FIG. 2 and FIG. 5 illustrate overhead wires for power supply including a trolley wire 201, a trolley wire 202, and a transverse support member or bracket 203.
[0007]
2 and 5 show two insulators, a hanger 206, a suspension wire rotary clamp member 503, a pointed clasp 504, a jib pipe 505, and a diagonal pipe 506.
[0008]
In the case of the method known from document [2], professionals visually monitor and evaluate the video images for trolley wire damage or foreign objects on the overhead wire.
[0009]
Further, another method of automatically monitoring a railway traveling route is known from Documents [3], [4], [5], and [6], respectively, or proposed in Document [8]. In this case, triangulation [3] and light propagation time based phase measurement [4] are used to determine the height and lateral position of the trolley line.
[0010]
To determine the wear of the trolley wire, the width of the sliding surface is measured using a laser scanner [5] or a diode array [2], or the residual height of the trolley wire in the horizontally oriented transmitted light is reduced. Required [6].
[0011]
The measuring device known from Ref. [6] used for this purpose is mounted on the top of the slider, which limits the vehicle speed for mechanical reasons and causes shadows in the overlapping area of the trolley wires. This may result in the erroneous recognition that a residual height greater than the actual height exists. In order to prevent this, such a region is excluded in the measurement.
[0012]
A further disadvantage of the recording unit-based contactless monitoring method with a substantially vertical gaze direction of the recording unit is that it can only capture the width of the mirror surface. In the case of a trolley wire having a substantially rectangular profile, the residual height cannot be determined even by a mirror surface.
[0013]
Other components of the current supply, such as the longitudinal link and the transverse support, cannot be checked automatically by known monitoring methods. Therefore, in general, a video camera is used to record the current supply section. However, since the resolution and the image frequency of the video camera are considerably low, only very rough defects can be identified. After the inspection run, all video images must be visually monitored and evaluated by a specialist or inspection personnel.
[0014]
An alternative would be for inspection personnel to patrol the runway, which would be very time consuming, costly, and involved the risk of some accident.
[0015]
It is therefore an object of the present invention to be able to automatically and easily detect changes in objects on a roadway, thereby making the roadway simpler and less expensive and more expensive than the known methods described above. An object of the present invention is to provide a method and an apparatus capable of performing monitoring with reliability.
[0016]
BRIEF DESCRIPTION OF THE INVENTION
This problem is solved by a method and a device according to the features of the independent claims.
[0017]
According to the invention, in a method for determining a change of a reference object belonging to a travel path and provided at a predetermined location in the travel path, the following steps are performed:
Recording at least one reference image at a first point in time at a predetermined location of the travel path;
Forming an image of a reference object in said at least one reference image;
Recording at least one inspection image at a second point in time at a predetermined location of said travel path;
Forming another image of a reference object in said at least one inspection image
A step of comparing the image of the reference object of the at least one reference image with another image of the reference object of the at least one inspection image to determine a change of the reference object.
[0018]
Further according to the invention, in a device for determining a change of a reference object belonging to a travel path and provided at a predetermined location in the travel path, the following means are provided:
A first recording unit for recording at least one reference image at a first point in time at a predetermined location of said travel path;
A first evaluation unit for forming an image of a reference object in the at least one reference image
A second recording unit for recording at least one inspection image at a second time at a predetermined location of said travel path;
A second evaluation unit for forming another image of a reference object in the at least one inspection image
A third evaluation unit for comparing the image of the reference object of the at least one reference image with another image of the reference object of the at least one inspection image to determine a change of the reference object.
[0019]
This device is suitable, for example, for implementing the method according to the invention or its embodiments described below.
[0020]
The dependent claims show advantageous embodiments of the invention.
[0021]
The present invention and the embodiments of the present invention described below can be realized as software, or can be realized as hardware using a dedicated electric circuit.
[0022]
Furthermore, the present invention and its embodiments can be realized by a storage medium readable by a computer, and the storage medium stores a computer program for executing the present invention or the embodiments.
[0023]
Further, the present invention or each embodiment described later can be realized by a computer program product. The computer program product includes a storage medium storing a computer program for executing the present invention or the embodiment. ing.
[0024]
By recording a plurality of reference images and / or a plurality of inspection images at a predetermined location, it is possible to increase the reliability in determining a change in the reference object. In this case, it is preferable to record a plurality of reference images using different recording parameters such as recording angles. The same is true for inspection images.
[0025]
If the reference image and the inspection image belonging to the reference image are recorded using the same recording parameters, the reliability can be further improved.
[0026]
Advantageously, in order to reduce the storage space and the complexity of the calculation, a first partial image in the reference image and / or a second partial image in the inspection image are formed, which may comprise an image or another image. are doing. In this case, the first partial image and the second partial image are compared with each other for comparison.
[0027]
It is further useful to arrange the first partial image in the reference image and / or the second partial image in the inspection image as follows. That is, the first partial image and / or the second partial image are arranged so as to have a periphery belonging to the reference object.
[0028]
According to one embodiment, one partial image is a search mask. Using this search mask, a determined image area can be examined for an image or another image or change.
[0029]
The comparison between the first partial image and the second partial image can be easily realized as follows. That is, the first pixel value of the first pixel in the first partial image is subtracted from the second pixel value of the second pixel in the second partial image. Such a first pixel value and such a second pixel value can be, for example, a luminance value or a color value (luminance value or chrominance value).
[0030]
By subtracting the pixel values in this manner, a difference image is formed, and in this difference image, an image change between one image and another image can be easily obtained. Then, the change of the reference object can be obtained using the image change.
[0031]
Further, the difference image can be further processed, for example, by filtering and / or segmenting the difference image. Thereby, the reliability at the time of detecting the change can be improved.
[0032]
According to one embodiment, an image of the reference object in the reference image or another image of the reference object is determined using an edge detection method or an object identification method. Such a method is known from references [10] and [11].
[0033]
Another image of the reference object in the inspection image can be determined very easily by using a correlation method as known from references [9], [12], [13]. In this case, a correlation between a pixel value belonging to one image and a pixel value belonging to another image is obtained.
[0034]
The reliability in determining the change of the reference object can be further increased as follows. That is, the first magnitude of the geometric reference characteristic of the reference object is determined using the image of the reference object in the reference image, and the same magnitude in the reference object is determined using another image of the reference object in the inspection image. The second magnitude of the geometric reference characteristic is determined.
[0035]
For example, the angle of the reference object, the length of the reference object, the curvature of the reference object, the interval of the reference object, or the shape of the reference object can be determined as the geometric reference characteristic.
[0036]
In this case, the first size is compared with the second size in the comparison. In this comparison, a difference between the first size and the second size can be formed, and a change can be detected when the difference exceeds a threshold.
[0037]
According to one embodiment, the reference image and the inspection image are associated using the location code and / or the time code. Such a location code can be, for example, a number assigned to the reference image and / or the inspection image. In addition, this type of time code can be, for example, time information assigned to a reference image and / or an inspection image.
[0038]
According to one embodiment, a plurality of locations are preset. The method according to the invention is carried out at each of these preset locations.
[0039]
According to one embodiment, the reference object is decomposed into a plurality of reference partial objects.
[0040]
According to one embodiment, the reference object is a functional component of the roadway, for example a traffic technology object. In this case, the travel path can be a traffic communication network, for example, a railway line section.
[0041]
According to yet another embodiment, the reference object is a member of a track. This type of traveling track can be used as an overhead line. In this case, the reference object is a member of the longitudinal chain mechanism on the catenary, i.e., a member of the trolley, or a member of the transverse support member, i.e., the bracket, on the catenary.
[0042]
Also, one of the reference objects or the reference partial objects can be a basic member of the track, such as a rod or pipe or a cable or a wire, or a connecting member of the track, such as a holding member or a clamping member or a supporting member or an insulator. be able to.
[0043]
According to one embodiment, the reference object is a member of a rail stretch.
[0044]
By using each of the first recording unit and / or the second recording unit as a digital camera, a particularly simple device according to the present invention can be realized. Also, the evaluation unit can be realized very easily by a PC, and a computer program capable of executing the corresponding steps is stored in the PC.
[0045]
According to one embodiment, which is used for monitoring a running path, a vehicle traveling on a running path is equipped with the above-described device.
[0046]
BRIEF DESCRIPTION OF THE DRAWINGS The drawings show an embodiment of the invention, which will now be described in detail.
[0047]
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a diagram showing a method of performing image evaluation by identifying a change in an overhead wire object.
[0048]
FIG. 2 is a diagram illustrating overhead lines of a railway.
[0049]
FIG. 3 is a diagram illustrating an inspection vehicle according to the embodiment.
[0050]
FIG. 4 is a diagram showing a recording system for recording the overhead wire.
[0051]
FIG. 5 is a view showing a visual field for inspecting the bracket.
[0052]
FIG. 6 is a diagram showing the setting of the evaluation window.
[0053]
7A and 7B are diagrams illustrating a method of inspecting an overhead wire according to one embodiment.
[0054]
FIG. 8 is a diagram illustrating an overhead wire inspection method according to one embodiment.
[0055]
FIG. 9 is a diagram showing an image representing the determination of the search mask.
[0056]
Example: Optical automatic inspection of overhead lines of railway communication network
System concept of optical automatic inspection
FIG. 3 shows an inspection vehicle 300, which is a locomotive modified for railroad section inspection.
[0057]
The purpose of the inspection is to detect changes in overhead lines in railway tracks. Such a change is caused, for example, by damage occurring in a member of the overhead wire or foreign matter entangled in the overhead wire.
[0058]
If a change is detected and classified as having a decisive effect on the unhindered operation of the railway track, measures are taken to eliminate the change. Such measures may include, for example, repairing damage or removing foreign matter.
[0059]
It should be noted that the inspection system described below can be used not only for the inspection of the overhead lines of the railway communication network, but also for the inspection of any components in the railway communication network with appropriate modification. For example, it can also be used for inspection of rail bodies and rail sleepers.
[0060]
A plurality of optical recording systems 302, 303, 304, 305 are provided on a roof 301 of the inspection vehicle 300, and these optical recording systems are used during the recording traveling of the inspection vehicle 300 to set a predetermined recording location 300. Is recorded in a contactless manner from various directions with respect to one location without contact.
[0061]
These optical recording systems 302, 303, 304, 305 are oriented so that in the images recorded by these systems, each corresponding to one location, all members of the catenary line at that location and their surfaces can be identified. Have been.
[0062]
The optical recording systems 302 and 303 are used for monitoring and inspection of the transverse support member 306, the optical recording system 304 is used for monitoring and inspection of the overhead wire 307, and the optical recording system 305 is used for monitoring and inspection of the trolley wire 308. Used for inspection.
[0063]
In FIG. 2, one recording location is illustrated by way of example, along with an overhead wire and overhead wire members, such as a bow wire 310 (see FIG. 2) and a transverse support member 311 (see FIG. 2).
[0064]
Since all the images to be recorded are encoded and stored for the location and time, each image can be uniquely associated with a recording location belonging to it and a recording time belonging to it.
[0065]
After the inspection run, the stored images are evaluated using an automatic image processing method described later.
[0066]
In this evaluation, the image of the test run is compared with a reference image recorded at a previous point in time at the same predetermined location under the same recording parameters. The reference image is associated with the inspection image belonging to the reference image by using the location code and the time code assigned to the image.
[0067]
In this comparison, changes in the catenary line or members of the catenary line are detected, which may be recorded in certain protocols and automatically classified into one of three classes. Here, the three classes are defined defects, unclear situations, and minor differences. Inspection personnel will evaluate only unclear situations later. Then, for the detected change, maintenance instructions or repair measures are ordered as necessary.
[0068]
Inspection personnel need only visually check for unclear situations, thus freeing time-consuming and cumbersome inspection of a much larger area without problems.
[0069]
The resolution of the recording system 302, 303, 304, 305 needs to be approximately 1-2 mm to be able to identify relatively small members in the overhead line, such as bolt heads or individual wires jumping out of defective cables. . Also, the exposure time of the recording system 302, 303, 304, 305 should not be longer than about 45 ms to avoid intensifying motion blur in the image due to a vehicle speed of 80 km / h. . As a result, the amount of data to be stored is 100 Mbytes / s or 5 Gbytes / km. Detailed specifications of the recording system will be described later.
[0070]
In the case of the friction measurement of the trolley wire 308, it is not necessary to compare the inspection image with the preceding reference image. In this case, it is sufficient to compare the profile of the trolley wire 308 recorded during the inspection run with a predetermined target profile. The measurement inaccuracy of the corresponding recording system 305 must be about 0.1 mm, and it must be possible to identify profile constrictions of 2-3 cm.
[0071]
Since the duration of a single measurement for the overhead wire 307 does not exceed 45 ms, and the duration of a single measurement for the trolley wire profile measurement does not exceed 200 ms, the vibration of the inspection vehicle 300 or the trolley wire 308 less than 5000 Hz is a result of the measurement. Does not affect the quality of the product. For higher frequencies the amplitude is smaller than the required resolution and therefore has no disturbing effect.
[0072]
Automated optical inspection system
The device depicted in FIG. 4 is used to record the overhead wire 307 or 400. On the left side 401 and right side 402 of the vehicle ceiling 403, one diode array camera 404, 405 is provided together with the illuminations 406, 407 belonging thereto.
[0073]
Each diode array camera 404, 405 records a vertical image stripe on one side of the overhead wire 400 within 45ms. In this case, an image that extends “infinitely” occurs in cooperation with the horizontal movement of the inspection vehicle 300. At a vehicle speed of 80 km / h, 22,000 lines are recorded per second, and each stripe is 1 mm wide. The camera array has 2048 elements, which at the object to be recorded corresponds to a resolution of 0.7 to 2 mm in the direction of the camera array, depending on the distance.
[0074]
The overhead wire 400 is illuminated for the purpose of realizing use at night and obtaining a prescribed light condition during the day. As the lightings 406, 407, metal vapor projectors 406, 47 having a connection load of about 10 kW for the moment were considered. However, since it is desired to avoid dazzling due to unrelated illumination due to a space problem caused by providing a plurality of cameras, an infrared diode laser was used. This illuminates the field of view of the individual camera more accurately than a metal vapor projector and does not illuminate extraneous areas, so a power of 15 W is sufficient.
[0075]
Further, in order to prevent human eyes from being exposed to danger, about 100 lenses are arranged as a linear raster having a length of about 30 cm at the emission opening of the laser casing of the infrared diode lasers 406 and 407. This allows as many as 100 side-by-side lasers to be seen, rather than the strong point lasers. The irradiation intensity generated by the infrared diode lasers 406 and 407 is linearly attenuated as the interval becomes wider because the opening angle of the emitted laser beam is 40 ° (see FIG. 4). Therefore, the maximum exposure time of the pupil is 2 ms. When it is less than that, ie when the inspection vehicle 300 exceeds the minimum speed of 20 km / h, the danger is eliminated according to internationally valid laser beam protection regulations already from a minimum distance of about 2 m.
[0076]
In the case of laser irradiation, in general, disturbing granulation occurs in an image due to interference. In this case, such an obstacle is completely avoided. The reason for this is that the infrared diode lasers 406 and 407 used have a plurality of light sources that produce overlapping light without interference, unlike a single light projector.
[0077]
FIG. 5 shows a field of view 500 of a diode array camera for recording the bracket 501. Since this diode array camera approaches or moves away from the bracket 501 and does not move parallel to the target object as in the case of a catenary, it is positioned horizontally to record a two-dimensional image. It is necessary to move the field of view 500 of the diode array camera vertically using a rotating mirror.
[0078]
The resolution of the diode array camera is 2 mm in the horizontal direction and about 1 mm in the vertical direction due to oversampling. Since 180 ms is required to scan 4048 lines at a frequency of 22000 lines per second, the distance between the diode array camera and the bracket 501 changes by about 4 m based on the movement of the inspection vehicle.
[0079]
For example, an image failure that may occur due to a swing of the inspection vehicle or the like can be compensated for by a swing sensor by using continuously recorded data.
[0080]
A laser radar system 322, 323 [6] is provided on the ceiling 301 near both end surfaces 320, 321 of the inspection vehicle 300, respectively, and the beam of this system is oriented vertically upward. The trolley wire side holding member of the transverse support member 306 is thereby identified.
[0081]
As soon as the laser radar system 322 ahead in the direction of travel identifies the transverse support member 306, the system triggers image recording by the scanner camera 324 pointing in the direction of travel. Passing the transverse support member 306 triggers the rearward facing scanner camera 325. In this manner, the object can be recorded from both sides.
[0082]
FIG. 3 shows the beam fans 331, 332, and 333 of the recording systems 302, 303, and 304. For the sake of clarity, only one beam fan 333 is illustrated for the beam fan of the overhead wire 307. This beam fan is generated almost vertically on the plane of the overhead wire 307, while the beam fans 331 and 332 for the transverse support member 306 are arranged almost horizontally. Ambient influences that impair the measurement can be minimized by:
A narrow band filter placed in front of the camera allows only light having the wavelength of the aforementioned infrared laser to pass, so that measurements can be made during the day as well as at night. In this way, sunlight can be suppressed as long as it does not hit the camera directly.
[0083]
Since the temperature on the roof can fluctuate more than the laser, camera and optics allow, the interior of all casings is sealed and circulated with water to a temperature of about 20 ° C. Is done.
[0084]
The window in front of the optics ensures that it is protected against light rain during the measurement. In the case of very heavy rain, fog or snowfall, the visibility of the transverse member at a distance of more than 10 m can be obstructed, and in some cases good measurement results can no longer be obtained.
[0085]
Cleaning of windows for dirt such as dust and soot can be automatically performed by hot water injected using pressure. Therefore, there is no need to climb the ceiling of the vehicle. The window is closed by the protective cover when no measurements are taken.
[0086]
Image rating
FIG. 1 illustrates the principle of an image evaluation 100 during the automatic optical inspection of overhead lines in a railway communication network, which will now be described.
[0087]
An image (inspection image) 101 recorded during the inspection traveling is transmitted to a computer (industrial PC) located inside the inspection vehicle. First, the inspection image 102 is compressed to about 3% of the original data capacity (103), buffered on a hard disk, and then sent to a magnetic tape.
[0088]
This type of magnetic tape can store all data of a single line section of about 300 km. Since there is a hard disk that acts as a buffer, there is no need to interrupt measurement when changing tapes.
[0089]
After the test run, the automatic evaluation 110 can be performed during a relatively long maintenance downtime or during a run without measurement. The introduction of additional computing techniques can speed up the onset of evaluation 110 on the vehicle.
[0090]
The prerequisite for the image evaluation 110 is that the so-called reference travel is performed at the time of the preceding travel. In this reference travel, the actual state (reference state) is recorded by the reference image 102, which is determined and the reference data is determined. Is formed.
[0091]
The reference image 102 and the inspection image 101 are identical in their recording locations and recording parameters.
[0092]
FIG. 1 further shows a comparison (120) between the reference image 102 from the reference run and the corresponding test image 101 from the current test run.
[0093]
Three methods for identifying changes in the catenary line are performed during image evaluation 110 or 130, and these methods will now be described for identifying changes in the transverse support members and changes in the catenary line:
a) Method 131 based on geometry comparison 131
b) Difference image based method 132
c) Foreign matter identification method 133
Basic principles of a) to c)
Each of these three methods 131, 132, 133 is based on a comparison 130 of the reference image 102 and the corresponding inspection image 101. At the time of this comparison 130 or 135, an image 136 of the object selected in the reference image is first formed. Then, in the corresponding inspection image, an image 137 of this selected object is also formed.
[0094]
A change in the selected object is determined 140 by comparison 135 of both images 136, 137 of the selected object or by comparison of measurements determined using the images.
[0095]
For methods a) and b):
FIG. 6 shows the format and structure of reference data relating to the overhead lines 601 and 602. These data are derived from a large number of very large overall images, which have overhead lines 601 and 602 between the two transverse support members 603 and 604. The reference data includes, for example, small partial images 611, 612, 613, and 614 that image components such as the clamp members 621 and 622 in a target state, their geometric dimensions and relationships, and positional information related to sections. include. The cable 631 and the trolley wire 632 are defined by their diameter and the fixation points 633, 634 located in the partial image.
[0096]
The automatic identification of the transverse support members 603, 604 and the continuous recording of the distance counter pulse guarantees a posterior correspondence of the recording location more than 0.5 m. Since the information regarding the recording location exists during the automatic comparison, the computer can easily find the images related to the reference travel and the test travel by the computer.
[0097]
FIG. 7a shows a reference image 701 of a partial area of the transverse support member 702 (without trolley line and support rope). The partial images or the evaluation windows 703, 704, 705, 706, 707 are represented by thin lines.
[0098]
FIG. 7b shows the corresponding inspection image 710. However, before this recording, the positions of the clamp member 720 (reference image) and 721 (inspection image) are shifted, and the side end of the rope 722 protruding sideways is inserted as a defect. A change generated by the image processing is found and sent as a notification to the defect record.
[0099]
The displacement of the clamping members 720 or 721 causes a change in the spacing 724 and the angle 725, so that the laterally protruding rope end 722 can be clearly identified. In the same way, other changes in other components can be found.
[0100]
Method c):
FIG. 8 shows a partial region of the transverse support member 801 in the inspection image 800.
[0101]
In the inspection image, two search masks 802 and 803 are drawn, which are defined in advance in the corresponding reference image.
[0102]
In the reference image these search masks are generally defined between the objects shown so that they cover the image area, the so-called search area. By doing so, it is possible to identify a foreign substance by comparing the search area in the reference image with the corresponding search area in the inspection image.
[0103]
Specifically, in this method, a search area based on the reference image is applied to the inspection image (after being adjusted to a change in the position of an object in the inspection image as necessary), and at that time, In the search area, a search is performed for a foreign substance that is generally distinguished due to luminance on a dark background.
[0104]
FIG. 8 shows the unraveled end 810, which is identified by comparing the search area 802 of the inspection image 800 with the corresponding search area in the associated reference image.
[0105]
In this way, it is possible to easily identify an extra object, such as a foreign substance, which is imaged in the inspection image but not in the reference image.
[0106]
The determination of the search mask is performed as follows (FIG. 9):
First, a reference image 900 having various objects 901, 902, 903 and having their pixels having various gray values 904, 905 is selected and binarized (910).
[0107]
The objects 901, 902, and 903 appear white in the binary reference image 911. Other areas appear black.
[0108]
In the following step 920, the width of the white area is increased (image 921).
[0109]
At that time, the search areas 932, 933, and 934 are determined such that the edges of the associated black areas are surrounded by the search mask. The various parts of the search mask are stored together in a single binary search mask image.
[0110]
The following describes an alternative embodiment to this example:
In this alternative embodiment, a location coding method for associating a reference image with an inspection image will be described.
[0111]
In order to be able to assign the corresponding co-located reference image to the inspection image, the location information from the distance measuring system is continuously included in each image, i.e. both the reference image and the inspection image, during each recording run. Recorded in. In addition to this, a special system for identifying the main object (for example the transverse support of an electric railway) records the presence of the transverse support or the strut together with the corresponding location information.
[0112]
This allows the transverse support members, ie their struts, to be counted and numbered along the running section. These numbers are used for associating an appropriate reference image with a corresponding inspection image upon inspection.
[0113]
The following publications have been cited herein:
[1] ISO / IEC JTC1 / SC29 / WG11, MPEG-4 Video Verification Model Version 5.0 Doc. N1469, 11. 1996, p. 55-59
[2] M. Van Gigch, C.I. Smallenburg, A.M. W. 3. Benshop, "System zur Message der Fahrdrahtdicke (ATON) der Niederlaendischien Bahnen", Schienen der Welt, 4. 1991, p. 20-31
[3] B. Sarnes, "Qualitaetssicherung an Stromabnehmer und Oberleitung", ETR, 48 (1999) 3, p. 117-123
[4] H. Hoefler, N .; Dimopoulos, B .; Metzger, "Fahrdrahtlagemessung auf Hochgeschwindigkititsstrecken", eb, 96 (1998) 3, p. 61-64
[5] S.P. Ishizu, "Series 700 derivative will replace Doctor Yellow", Railway Gazette Intern. , 4. 1999, p. 218-220
[6] P.I. 5. Pohl "Der Hagener Video-Mestriebwagen", Der Eisenbahningenieur, 6. & 7. 1996, Vol. 6 & 7 1996, p. 24/32
[7] R.I. Mueller, H .; Hoefler, "Fahrwegweberwachung mit opscher Messtechnik", Eisenbahn-Ingenieur Kalender '97, Darmstadt: Tetzlaff, 1996, p. 315-332
[8] DE 199 36 448 A1
[9] Available from September 1, 2000 below
http: // www. it. kth. se / docs / iso-13818-2. read. html
[10] Available on or after September 1, 2000 at
http: // vvl1. mit. fh-heilbrown. de / German / Theorie / Kantendetection. htm
[11] Available from September 1, 2000 on
http: // www. ifi. unith. ch / groups / mml / people / mbichsel / Voresung. html
[12] B. Jaehne, Digital Bildverarbeitung, Springer Verl. 1997
[13] F.I. M. Wahl, Digital Bildsignalverarbeitung, Kapite 6.5, Detektionsfilter, Springer Verl. 1984
[Brief description of the drawings]
FIG.
It is a figure which shows the method of performing image evaluation by discriminating the change of an overhead wire object.
FIG. 2
It is a figure which shows the overhead line of a railroad.
FIG. 3
It is a figure showing an inspection vehicle by an example.
FIG. 4
It is a figure which shows the recording system for recording a trolley wire.
FIG. 5
FIG. 3 shows a field of view for inspecting a transverse support member.
FIG. 6
It is a figure showing setting of an evaluation window.
FIG. 7
It is a figure showing the overhead wire inspection method by one Example.
FIG. 8
It is a figure showing the overhead wire inspection method by one Example.
FIG. 9
FIG. 9 is a diagram illustrating an image showing how a search mask is determined.

Claims (29)

A method for determining a change in a reference object belonging to a travel path and provided at a predetermined location in the travel path,
Recording at least one reference image at a first point in time at a predetermined location of the travel path;
Forming an image of a reference object in the at least one reference image;
Recording at least one inspection image at a second point in time at a predetermined location of the travel path;
Forming another image of a reference object in the at least one inspection image;
Comparing the image of the reference object of the at least one reference image with another image of the reference object of the at least one inspection image to determine a change in the reference object,
A method for determining the change of a reference object. The method of claim 1, wherein a plurality of reference images are recorded at predetermined locations. 3. The method according to claim 1, wherein a plurality of inspection images are recorded at a predetermined location. 4. The method according to claim 1, further comprising forming a first partial image having an image in the reference image. 5. The method according to claim 1, further comprising forming a second partial image having another image in the inspection image. 6. The method according to claim 5, wherein the comparison compares the first partial image with the second partial image. 7. The method according to claim 6, wherein, in the comparison using the first partial image and the second partial image, a difference image is obtained by subtracting a pixel value of the first partial image and a pixel value of the second image. The method of claim 7, wherein an image change between one image and the other image in the difference image is determined. 9. The method of claim 8, wherein the change in the reference object is determined using the change in the image. A first magnitude of the geometric reference characteristic of the reference object is determined using an image of the reference object in the reference image, and a second magnitude of the same geometric reference characteristic is determined using another image of the reference object in the inspection image. The method according to claim 1, wherein the magnitude is determined. The method according to claim 10, wherein an angle of the reference object, a length of the reference object, a curvature of the reference object, a spacing of the reference object, or a shape of the reference object is determined as the geometric reference characteristic. The method of claim 10 or 11, wherein the comparing compares a first magnitude with a second magnitude. 13. The method of claim 12, forming a difference between the first magnitude and the second magnitude, and determining that a change has occurred if the difference exceeds a threshold. 14. The method according to claim 1, wherein the reference image and the inspection image are associated with each other using a location code or a time code. 15. The method according to any one of the preceding claims, wherein a plurality of locations are predefined and the method according to any one of the preceding claims is performed at each of the predefined locations. 16. A method as claimed in any one of the preceding claims, wherein the method is applied for roadway monitoring and measures are taken to determine changes and remove the changes. An apparatus for determining a change in a reference object belonging to a travel path and provided at a predetermined location in the travel path,
A first recording unit that records at least one reference image at a first point in time at a predetermined location of the travel path;
A first evaluation unit for forming an image of a reference object in the at least one reference image;
A second recording unit for recording at least one inspection image at a second point in time at a predetermined location of the travel path;
A second evaluation unit for forming another image of a reference object in the at least one inspection image;
A third evaluation unit is provided for comparing the image of the reference object of the at least one reference image with another image of the reference object of the at least one inspection image to determine a change in the reference object. And
A device that determines the change of a reference object. The apparatus of claim 17, wherein the reference object comprises a plurality of reference sub-objects. 19. The device according to claim 17, wherein the reference object is a functional component of a traveling path. 20. Apparatus according to any one of claims 17 to 19, wherein the reference object is a traffic technology object. 21. Apparatus according to any one of claims 17 to 20, wherein the travel path is a traffic network. 23. Apparatus according to any one of claims 17 to 22, wherein the traffic network is a railroad section. 23. Apparatus according to any of claims 17 to 22, wherein the reference object is a track member. 24. The apparatus of claim 23, wherein the track is an overhead line. 25. The apparatus of claim 24, wherein the reference object is a member in a longitudinal chain member of the catenary or a member in a transverse support member of the catenary. One reference part object of the plurality of reference part objects is a basic member of a traveling line such as a rod or a pipe or a cable or a wire, or a connecting member of a traveling line such as a holding member or a clamping member or a supporting member or an insulator. Apparatus according to any one of claims 18 to 25. 23. Apparatus according to any one of claims 17 to 22, wherein the reference object is a member of a rail stretch. 28. Apparatus according to any one of claims 17 to 27, wherein the first recording unit and / or the second recording unit are each a digital camera. 29. Apparatus according to any one of claims 17 to 28, wherein an apparatus adapted for roadway monitoring and performing the action of determining a change and removing the change is attached to a vehicle traveling on the roadway.

Images