KR102346438B1 - Image processing based concrete railroad track inspection apparatus and concrete railroad track inspection method using the same - Google Patents

Abstract

The present invention relates to crack detection of a railway concrete ballast 10 using image information processing of a computer 50, and is mounted on a bogie 20 running along a railway track to be inspected. The camera 30 for photographing (10) is connected to the computer 50, and the computer 50 binarizes the image information so that the location and shape of the crack can be accurately identified.
Through the present invention, a safe, simple and precise inspection of the concrete railway ballast 10 is possible, thereby remarkably improving the convenience and accuracy of the railway ballast 10 inspection.

Description

Image information-based railway concrete ballast crack detection device and method using the same

The present invention relates to crack detection of a railway concrete ballast 10 using image information processing of a computer 50, and is mounted on a bogie 20 running along a railway track to be inspected. The camera 30 for photographing (10) is connected to the computer 50, and the computer 50 binarizes the image information so that the location and shape of the crack can be accurately identified.

Railroads have advantages in terms of transport capacity and efficiency compared to other land transport methods.

Among railway facilities, the concrete ballast is a structure that directly supports the load acting on the track. As with other concrete structures, the key inspection item is crack inspection.

The traditional method of inspecting cracks in concrete structures is a method of estimating the progress of cracks by periodically taking pictures of cracks and identifying the occurrence and shape of cracks by visually observing the structures to be inspected by a professional engineer. Field visits and short-distance access to the objects to be inspected are inevitable.

However, not only is the extension of the railroad track very long, but there are considerable difficulties in direct access and inspection by inspectors due to the risk of accidents. there was no

Accordingly, technologies such as Unexamined Patent Publication No. 2003-83359, which introduced a computer image processing technique to the crack inspection of concrete structures to promote the speed and convenience of inspection, have been developed and utilized, so that the inspector does not need close-range approach and direct visual observation. Rapid detection of cracks in structures has become possible through computer image processing.

Conventional computer image processing applied concrete structure crack inspection technology, including Patent Publication No. 2003-83359, basically acquires an image that is a digital image by photographing the surface of a concrete structure, and processes the obtained image using a computer algorithm to take the corresponding photographing. By judging whether or not there is a crack in a part, a numerical value that uniformly increases or decreases from the bright part to the dark part according to the brightness and darkness of the captured image information obtained from the camera is given, and it is binarized. ) to detect dark areas that contrast sharply with the background, which is mainly a bright area.

That is, by setting a numerical value of 0 to 255 commonly used as a gray scale for each pixel of the raw image information obtained through the camera, the darkest part and the brightest part are set. By assigning 0 and 225, respectively, the brightness of each pixel is digitized, and then binarization is performed in which a single value is assigned to pixels above and below a certain reference value. For example, by averaging the grayscale values of all pixels After setting the reference value, the imaging information is processed in such a way that 0 is assigned to a pixel exceeding the reference value and 1 is assigned to a pixel lower than the reference value.

In the binarized imaging information, 1 is given to pixels in which cracks, which are dark areas that clearly contrast with the background, are photographed, and 0 is given to pixels in which the surface of non-cracked areas, which can be said to be relatively bright, is photographed. Through this, the computer can grasp the occurrence and shape of cracks.

On the other hand, in the binarization process, 1 and 0, which are the given numerical values for each pixel exemplified above, are only logical values given for convenience in computational processing in bisecting information, and are equivalent to the numerical values 0 to 225 set in general grayscale. Instead of 1 and 0, various numbers such as 0 and 225 may be applied.

Through the conventional crack inspection technology for concrete structures applied with computer image processing, including the aforementioned Patent Publication No. 2003-83359, it was possible to secure the speed, convenience, and safety of crack inspection. It is only intended to identify cracks, and it is impossible to systematically record the exact location and specific shape of individual cracks.

In the prior art, the captured image is processed as available image information by applying the above-described binarization technique, thereby extracting the existence of a crack and the approximate shape of the crack. The exact location of the crack could not be determined only by the location of the crack.

Therefore, in the prior art, image information-based crack inspection is performed periodically, but once the image information for which the existence of cracks is identified is stored for each inspection period, the image information estimated to the same location is visually inspected to progress cracks. I had no choice but to check whether

That is, the image taken at intervals of several days to several months is compared with the naked eye afterward, or the crack location on the screen is adjusted by manpower to superimpose the image information for each shooting point, and then the progress of the crack is grasped.

Therefore, in the prior art, it was possible to automatically determine whether a crack exists in a captured image through a computer image processing technique. There was a limitation that post-processing or processing by manpower was necessarily required.

The present invention was devised to accurately set the actual location of cracks in a captured image in consideration of the above-described problems, and a bogie ( 20) is collimated downward and a camera 30 for photographing the railway map 10 is mounted, and the camera 30 is connected to the computer 50 so that the imaging information of the camera 30 is input to the computer 50 It is an image information-based railway concrete ballast crack detection device.

In addition, the camera 30 and the bogie 20 are connected by a connecting device 31 , the connecting device 31 having a connecting rod 32 having one end joined to the vehicle body 21 and protruding to the front of the bogie 20 and , a vertical rotating rod 33, which is a vertical rod, the upper end of which is connected to the front end of the connecting rod 32, and a horizontal rotating rod 34 whose middle is connected to the lower end of the vertical rotating rod 33 as a rod parallel to the connecting rod 32 and , a bar body orthogonal to the horizontal rotating rod 34 in plan view, the central portion of which is connected to the rear end of the horizontal rotating rod 34, and a lateral extension rod 35 coupled to both ends of the lateral extension rod 35, respectively, a counterweight (36) The front end of the connecting rod 32 and the upper end of the vertical rotating rod 33 are connected to be freely rotatable about the axis of rotation parallel to the connecting rod 32, and the lower end of the vertical rotating rod 33 and the horizontal rotating rod 34 It is an image information-based railway concrete ballast crack detection device, characterized in that the middle part of the connecting rod 32 is freely rotatably connected to the axis of the rotation axis orthogonal to the connecting rod 32, and the camera 30 is fastened to the front end of the horizontal rotation rod 34.

In addition, a sleeve 65 coupled to the vehicle body 21 of the bogie 20 so that a lifting rod 66, which is a vertical rod, can be freely raised and lowered is joined, and a driving wheel 61 running along a railroad track is mounted on one side. The mounted rotation system 62 is bonded to the lower end of the lifting rod 66, and the spring 67 is coupled to the lifting rod 66 between the sleeve 65 and the rotation system 62, so that the elastic force of the spring 67 is increased. An image characterized in that the driving wheel 61 is always in close contact with the railroad track by acting between the sleeve 65 and the rotation system 62, and the display 63 of the rotation system 62 is included in the imaging area of the camera 30 It is an information-based railway concrete ballast crack detection device.

In addition, in the method for detecting cracks in railway concrete ballast 10 using the image information-based railway concrete ballast 10 crack detection device, the camera 30 mounted on the bogie 20 is a marker M1 formed on the ballast 10 , M2) to be included in the photographing step (S10) of photographing the image 10, the input step of inputting the photographing information, which is raw information captured by the camera 30, to the computer 50 (S20), and the computer 50 ) receives the imaged information and stores it in a storage device (S30), and the computer 50 retrieves the stored imaged information and binarizes the imaged information to extract the markers M1 and M2 having a preset shape. Step S41, a calculation step S42 in which the computer 50 calculates the adjustment distance using the markers M1 and M2, and the computer 50 corrects and corrects the imaging information using the calculated adjustment distance It is characterized in that it consists of a correction step (S43) of generating information, and a recording step (S50) of the computer 50 binarizing the correction information to extract the crack (C), which is a dark part, and storing the binarized correction information in a memory device. It is an image information-based railway concrete ballast crack detection method.

In addition, the first light emitter 41 and the second light emitter 42 are mounted on the cart 20 , and in the photographing step S10 , the camera 30 moves the image 10 while the first light emitter 41 is turned on. ) to include the markers (M1, M2) formed in the image 10, and after the first image information is generated, the first light emitter 41 is turned off and the second light emitter 42 is turned on. The image 10 is photographed to include the markers M1 and M2 formed in 10 to generate second imaging information, and in the input step S20, the first imaging information that is raw information captured by the camera 30 and inputting the second sensing information into the computer 50, and in the storing step (S30), the computer 50 receives the first sensing information and the second sensing information and stores the received information in a storage device, and stores the first sensing information as the sensing information. The first correction information is generated by applying the imaging information to the positioning step S41, the calculating step S42, and the correction step S43, and the second imaging information is applied as the imaging information to the positioning step (S41). ), the calculation step (S42) and the correction step (S43) are performed to generate the second correction information, and in the recording step (S50), the first correction information and the second correction information are each binarized to form the first binary information and the second correction information. By generating binary information and calculating the logical values for each pixel of the first binary information and the second binary information in a logical product method to generate difference information, the dark part, the crack part (C), is extracted and the difference information is stored It is a method for detecting cracks in railway concrete ballast based on image information, characterized in that it is stored in the device.

Through the present invention, safe, simple and precise inspection of the concrete railway ballast 10 is possible, thereby remarkably improving the convenience and accuracy of the railway ballast 10 inspection.

In addition, since the actual location and absolute length of the crack can be accurately identified in the captured image, it is possible to accurately establish key information required for judging the integrity of the island 10 , such as whether or not the crack progresses and the speed of progression.

In particular, since the location specification of cracks as described above can be automatically performed through information included in the image, efficiency and speed in post-processing as well as inspection can be improved.

1 is a perspective view of the device of the present invention;
2 is an explanatory view of the photographing method of the present invention;
3 is an excerpted perspective view of the camera and the connecting device of the present invention
4 is a flowchart of the present invention;
5 is an exemplary view of image information of the present invention;
6 is an explanatory diagram of an image information processing process of the present invention;
7 is an excerpted perspective view of a multicolor light application embodiment of the present invention;
8 is an explanatory diagram of a correction information processing process in a multicolor light application embodiment of the present invention;
9 is a perspective view of a device according to an embodiment of the present invention
10 is an explanatory view of the shooting method of the mileage parallel side embodiment of the present invention
11 is an exemplary view of image information according to an embodiment of the mileage parallel side of the present invention

The detailed configuration of the present invention will be described with reference to the accompanying drawings.

First, FIG. 1 is a perspective view showing the external appearance of the apparatus of the present invention, that is, an image information-based railway concrete ballast 10 crack detection device. As shown, the ballast 10 crack detection device of the present invention runs along a railroad track. It consists of a bogie 20, a camera 30 mounted on the bogie 20, and a computer 50 connected to the camera 30, and the like.

The railroad bogie 20 is a non-powered track traveling device, and a plurality of independently freely rotating casters 22 are mounted on the lower portion of the vehicle body 21, which is usually composed of a rectangular frame, and is a self-propelled railroad vehicle or manpower. It is towed or propelled to travel along the railroad tracks.

As the bogie 20 applied to the present invention, a vehicle specialized for carrying out the present invention that is premised on image capturing can be applied, as well as a normal track, wiring or tunnel and bridge inspection and repair workbench 20 may be applied. have.

2 is a view illustrating the performance of the present invention, the bogie 20 is running along the rail built on the top of the concrete railway ballast 10 to be inspected, and the camera 30 is located on the front side of the bogie 20 It is mounted to photograph the upper surface of the image (10).

That is, as in FIGS. 1 and 2 , a plurality of casters 22 are mounted on the lower part of the vehicle body 21 and collimated downward to the bogie 20 moving along the railroad track is carried out. The camera 30 for photographing the image 10 is mounted, and the camera 30 is connected to the computer 50 so that imaging information of the camera 30 is input to the computer 50 .

Here, the computer 50 is mounted on the bogie 20 on which the camera 30 is mounted, as in the illustrated embodiment, so that the camera 30 and the computer 50 can be directly connected to each other through a wired/wireless connection method, as well as the computer ( 50) is not mounted on the same bogie 20 as the camera 30, but is mounted on a separate vehicle or possessed by an inspector, but the computer 50 and the camera 30 communicate with normal short-range wireless communication such as Bluetooth or wireless It may be connected through wireless communication of various standards, such as LAN wireless communication or mobile communication network-based wireless communication.

On the other hand, in the present invention, the vehicle body 21 and the camera 30 of the bogie 20 are not simply connected to each other so as to offset vibrations or fluctuations generated in the process of the bogie 20 traveling along a railroad track, but in Fig. 2 and 3, it is connected through the connecting device 31 so that the posture and collimation state of the camera 30 can be stably maintained despite the vibrations or fluctuations that are inevitably caused in the driving process of the bogie 20. .

The connecting device 31 of the present invention is configured such that a plurality of rods are connected to be rotatably multi-axis, and a high degree of balance can be maintained. As shown in FIGS. 2 and 3, the connecting device 31 has one end A connecting rod 32 joined to the vehicle body 21 and protruding forward of the bogie 20, a vertical rotating rod 33 which is a vertical rod having an upper end connected to the front end of the connecting rod 32, and a connecting rod 32 As a bar parallel to the horizontal rotating rod 34, the middle part of which is connected to the lower end of the vertical rotating rod 33, and as a rod orthogonal to the horizontal rotating rod 34 in plan view, the center is connected to the rear end of the horizontal rotating rod 34 It is composed of a lateral extension rod 35 to be used, and a counterweight 36 coupled to both ends of the lateral extension rod 35, respectively.

In particular, as shown in FIG. 3, the front end of the connecting rod 32 and the upper end of the vertical rotating rod 33 are connected to be freely rotatable about a rotational axis parallel to the connecting rod 32 as an axis, and the lower end of the vertical rotating rod 33 and The middle portion of the horizontal rotating rod 34 is connected so as to be freely rotatable about the axis of rotation orthogonal to the connecting rod 32, and the camera 30 is fastened to the front end of the horizontal rotating rod 34, so that the vehicle body (20) while driving ( 21), the posture of the camera 30 can be stably maintained even when the vertical vibration and the left and right vibrations occur.

In addition, as shown in FIG. 3 , the horizontal rotating rod 34 is configured to be able to adjust the length of the counterweight 36 installation side, so that it is possible to respond flexibly to a weight change due to the replacement of the camera 30 . .

The method for detecting cracks in railway concrete ballast 10 using the image information-based railway concrete ballast 10 crack detection device of the present invention is as shown in FIG. It starts with a photographing step (S10) of photographing the image 10 to include the markers M1 and M2 formed in the .

Here, the marker formed on the ballast 10 is a fixture having a certain shape, as shown in FIG. 5 , and includes fasteners installed on the ballast 10 to fix a rail, etc. Since these markers have a constant shape, it will be described later In image processing including binarization to be performed, easy extraction or identification of marker images is possible.

In addition, since the marker such as the rail fastener is a facility fixed to the map 10 , it is possible to determine the exact photographing position of the photographed image based on this.

It is difficult to determine the position of the image capturing through the markers M1 and M2 due to the nature of the present invention in which the camera 30 is mounted on the bogie 20 and the image is taken in a moving state. Therefore, it is required to solve this problem, and by moving the photographed image forward and backward with respect to the markers M1 and M2, the image information can be corrected to the desired exact photographing position.

That is, in the image as illustrated in FIG. 5 , two markers M1 and M2 were applied in a single image. This positioning situation is set as in-place shooting, and the actually captured image is corrected according to a process to be described later.

When the above-described photographing step (S10) is completed, the input step (S20) of inputting the photographing information, which is the raw information captured by the camera 30, to the computer 50 through wired/wireless communication (S20) is performed, and then the computer ( 50) as shown in the upper left of FIG. 6, the storage step (S30) of receiving and storing the captured image information in the storage device is performed, the captured image information can be safely recorded in the storage device of the computer (50).

Thereafter, the positioning step S41 of extracting the markers M1 and M2 having a preset shape by retrieving the stored imaging information and binarizing the imaging information by the computer 50 is performed, and in the embodiment illustrated in FIG. 6 , the marker As a fastener for fixing the rail to the ballast 10 was applied.

Subsequently, as shown in the middle part of FIG. 6 , a calculation step S42 in which the computer 50 calculates the adjustment distance using the markers M1 and M2 is performed. In the embodiment illustrated in FIG. 6 , the upper side Since the midpoint between the marker M1 and the lower marker M2 is biased toward the upper side of the screen, it is necessary to move the image information to the lower side in the subsequent processing.

In addition, in the embodiment shown in Fig. 6, the actual rail fasteners are configured on both the left and right sides of the screen, but it is possible to proceed without unreasonableness even if only one fastener is selected as a marker among them.

In the embodiment illustrated in FIG. 6 , the adjustment distance may be the difference between the marker center line indicated by the dashed-dotted line on the screen shown in the middle part of the drawing and the actual screen center line, and the distance is set as the adjustment distance to correct the position thereafter is used for

When the calculation step (S42) is completed, as shown in the lower left of FIG. 6 , a correction step (S43) in which the computer 50 corrects the imaging information using the calculated adjustment distance to generate the correction information is performed, thereby It is possible to acquire image information in the same condition and condition as if it was photographed in a still state at an exact location.

Thereafter, as in the lower right of FIG. 6 , the computer 50 binarizes the correction information to extract the crack part C, which is a dark part, and stores the binarized correction information in the memory of the computer 50 . By performing step S50, it is possible to accurately identify and record the location and form of cracks generated on the railway ballast 10, and after regularly performing the above-described process, by comparing the finally obtained correction information You can monitor the progress of the crack.

In particular, through the above-described adjustment distance introduction and position correction, the image information obtained at each inspection can be corrected to an accurate position, so the progress and speed of cracks can be determined only by simply superimposing each correction information finally obtained. It can be accurately grasped, and since this simple overlapping process is easy to be automatically performed by the computer 50, it is possible to quickly, accurately, and conveniently process a vast amount of inspection results.

On the other hand, FIGS. 7 and 8 show an embodiment to which the first light emitting device 41 and the second light emitting device 42 that emit lights of different colors are applied, and in this multicolor light application embodiment, the first light emitting device ( 41) and the second light emitter 42 can be mounted anywhere in the configuration of the bogie 20, such as the vehicle body 21 or the connecting device 31, as long as the light can be irradiated to the image 10, which is the imaging surface, in FIG. As in , installed together with the counterweight 36 at both ends of the lateral extension rod 35 of the connecting device 31, or these first light emitters 41 and the second light emitters 42 themselves are counterweights 36 By being installed so as to have the functions of , more efficient operation is possible.

That is, by configuring the first light emitter 41 and the second light emitter 42 together with the counterweight 36 or as the counterweight 36 in the connecting device 31, direct connection with the camera 30 is possible, so that a high-level In addition to easy synchronization control, the collimation line of the camera 30 and the illumination radiation lines of the first light emitter 41 and the second light emitter 42 are always kept parallel according to the balance maintenance action of the connecting device 31 . It is possible to maintain the stability of the lighting state.

The first light emitter 41 and the second light emitter 42 are lighting for photographing of the camera 30, and are lighting that sequentially emits multicolor light instead of monochromatic light, and the first light emitter 41 and the second light emitter 42 are Lights of different colors are irradiated onto the subject.

For example, the first light emitter 41 emits red light and the second light emitter 42 emits yellow light.交互) lights up.

The application of two-color illumination through the first light emitter 41 and the second light emitter 42 forms a foreign material part D, which is a kind of dark part, due to the deposition of contaminants or attachments on the object to be inspected 10. This is to prevent misjudgment as a crack (C). Foreign substances such as stains or attachments formed on the surface change the contrast or contrast with the surrounding area depending on the color of the light, whereas the crack area is related to the color of the light. This is based on the principle of always observing as a dark part without

In this embodiment of the application of the first light emitter 41 and the second light emitter 42 of the present invention, the lighting state of the first light emitter 41 and the second light emitter 42 for the same point in the state in which the bogie 20 is moved Since the shooting is performed for each of the lighting states of the first illuminator 41, the position difference between the image taken under the lighting of the first light emitter 41 and the image taken under the second light emitter 42 is unavoidable. It is not large, and since it is possible to correct the position to the correct position through the introduction of the above-described adjustment distance and performing the correction step (S43), it is possible to block the occurrence of an inspection error due to a positional deviation.

That is, in performing the above-described photographing step (S10), the camera 30 in the state in which the first light emitter 41 is turned on so that the markers M1 and M2 formed on the image 10 are included. After the first image information is generated by photographing, the first light emitting device 41 is turned off and the second light emitting device 42 is turned on. The map 10 includes the markers M1 and M2 formed on the map 10. ) to generate the second imaging information, the first light emitter 41 and the second light emitter 42 are turned on and off in sequence, and the camera 30 is photographed twice. It consists of a number of steps. However, this series of procedures is instantaneously performed by the electrically automatically controlled camera 30, the first light emitter 41, and the second light emitter 42, and is performed collectively in a short time that the inspector cannot feel it. Therefore, the stop of the bogie 20, which is a low-speed moving body, is not required and can be performed without unreasonableness even in a moving state.

The alternating light emission of the first light emitter 41 and the second light emitter 42 and the instantaneous photographing of the camera 30 in each light emission situation are performed by a control device built into the camera 30 or are performed with the camera 30 It can be performed by a control device that is separately configured and mounted on the bogie 20. Light emission of artificial lighting such as high-speed flash photography and the synchronization control of the camera 30 are widely used techniques in the general digital camera 30. , there is no specific limitation on the scope of the claims.

Then, in the input step (S20), the first imaging information and the second imaging information, which are raw information captured by the camera 30, are input to the computer 50, and in the storage step (S30), the computer 50 ) receives the first sensing information and the second sensing information and stores the received information in a memory device.

That is, under the operating conditions of the first light emitter 41 and the second light emitter 42, the first image capturing information and the second image capturing information are generated as the image is captured twice, and the above-described input for each of the image capturing information is performed. Steps S20 to S30 are performed.

Next, the first imaging information is applied as the imaging information and the positioning step S41, the calculating step S42, and the correction step S43 are performed to generate the first correction information, and the second imaging information as the imaging information is applied and the positioning step (S41), the calculation step (S42), and the correction step (S43) are performed, whereby the second correction information is generated.

In the subsequent recording step (S50), as shown in FIG. 8, the computer 50 binarizes the first correction information and the second correction information, respectively, to generate the first binary information and the second binary information, and the computer 50 ) calculates the logical values for each pixel of the first binary information and the second binary information in a logical product method to generate difference information to extract the crack part C, which is a dark part, and store the difference information in the computer 50 It is stored in the device, thereby enabling accurate detection of actual cracks even when there is a part that can be confused with cracks in appearance due to deposits of contaminants or attachments on the object to be inspected 10 .

When the binarization process for the first and second correction information as exemplified in the upper part of FIG. 8 is described, first, for each of the first correction information and the second correction information, the average value of brightness for all pixels is calculated. The first binary information and the second binary information are generated by dividing the pixels into two and displaying the upper and lower pixels, and the first binary information and the second binary information are exemplified in the middle part of FIG. 8 .

The above-mentioned red light and yellow light are respectively applied as lights emitted from the first light emitter 41 and the second light emitter 42, respectively, and contaminants such as mud are deposited on the concrete ballast 10 to form a foreign body (D) Assuming the situation, a large amount of red light emitted from the first light emitter 41 is reflected from the surface of the concrete ballast 10 of gray scale close to white and the mud contamination of red tone, and the light beam irradiated into the crack in the crack part C Since this is not reflected, a dark part is formed, whereas in the light emission condition of the second light emitter 42 emitting yellow light, only a large amount of reflection is made only on the surface of the concrete ballast 10, and in both the foreign material part (D) and the crack part (C). As the slight reflection is made, the foreign material part D forms a kind of dark part.

As a result, the foreign object part D is detected as a bright part under the first light emitter 41 light emission, whereas the foreign material part D is detected as a dark part under the second light emitter 42 light emission. It is to prevent misjudgment by treating only the part detected as a dark part as a crack part (C) regardless of the process.

If the above process is explained from the point of view of logical operation performed in computational processing, a logical value of 1 is assigned to a pixel to which brightness information less than or equal to the average value of brightness is given and treated as a dark part, and to a pixel to which brightness information exceeding the average value of brightness is assigned. After generating each of the first binary information and the second binary information in such a way that 0 is assigned as a logical value and treated as a list, the logical product of the same pixel logical values of these first binary information and the second binary information is logically multiplied ( ∧) That is, if the two logical values are the same, the logical value is maintained, and if they are different, the logical value is processed in such a way that 0 is assigned to generate difference information as illustrated in the lower part of FIG. 8 .

Therefore, only the crack part (C) is displayed in the difference information finally generated through binarization and difference processing, but the foreign substance part (D) is not displayed, thereby preventing misjudgment and dramatically improving the accuracy of inspection. .

On the other hand, accurate identification of the photographing location is possible through image processing or recognition of the above-mentioned markers (M1, M2), but it is possible to accurately determine the progress of the cracks, such as whether or not the cracks are enlarged. This is to accurately prepare for errors, and therefore, in order to determine the position on the route at the time in the inspection of a long-distance section, it is necessary to process such as accumulating the marker position measurements for the entire section.

Of course, it is possible to obtain a precise position on the route through the accumulation processing of the marker recognition information, but if an error occurs such as omission or overlap of some markers during the accumulation process, a serious error may occur in the overall check value. Therefore, although the precision is relatively low compared to the marker processing, a mileage measuring means is installed so that the location on the route can be immediately grasped.

9 to 11 show an embodiment in which a driving wheel 61 and a rotation system 62 are applied as mileage measuring means. A vertical rod 66, which is a vertical bar, can be freely raised and lowered on the vehicle body 21 of the bogie 20. The sleeve 65 coupled so as to be bonded is joined, and the rotation system 62 mounted on one side of the driving wheel 61 running along the railroad track is joined to the lower end of the lifting rod 66, and the sleeve 65 and the rotation system ( 62), a spring 67 is coupled to the lifting rod 66 between the springs 67, and the elastic force of the spring 67 acts between the sleeve 65 and the rotation system 62, so that the driving wheel 61 is always in close contact with the railroad track. .

9 and 10, the sleeve 65 is a tubular body coupled to the center so that the lifting rod 66 freely slides in the axial direction, and the lifting rod exposed to the upper side of the sleeve 65 ( At the upper end of the 66), a head having an enlarged diameter is formed to suppress separation and separation of the lifting rod 66 by the elastic force of the spring 67.

The rotation system 62 connected to the driving wheel 61 of the present invention is connected to the computer 50 by wire or wireless, so that the measurement value can be input to the computer 50 as well as, as shown in FIG. 11 , the imaging of the camera 30 It may be configured to include the display 63 of the rotation system 62 in the region.

In particular, as in FIG. 11 , the display 63 of the rotation system 62 is included in the imaging area of the camera 30 so that the display 63 of the rotation system 62 is photographed together with the captured image of the image 10 . , without measures such as electrical connection between the computer 50 and the rotation system 62 or between the camera 30 and the rotation system 62, it is possible to efficiently record the measured values of the rotation system 62 as well as the rotation system 62 It is possible to fundamentally prevent errors or errors that may occur during the transmission and storage of the measured values to the computer 50, as well as to fundamentally block the error or manipulation of the inspector that may occur during the work process, The reliability of the results can be improved.

10 : icon
20: bogie
21: body
22 : Caster
30 : camera
31: connection device
32: connecting rod
33: vertical rotating rod
34: horizontal rotating rod
35: lateral extension bar
36: counterweight
41: first light emitter
42: second light emitter
50: computer
61: driving wheel
62: rotation system
63: display
65: sleeve
66: lifting bar
67: spring
S10: shooting stage
S20: input stage
S30: Save step
S41: Positioning step
S42: Calculation step
S43: Calibration step
S50: Recording stage

Claims (5)

A plurality of casters 22 are mounted on the lower part of the vehicle body 21 and a camera 30 for photographing the railroad map 10 by collimating downward is mounted on the bogie 20 moving along the railroad track, and the camera 30 ) is an image information-based railway concrete ballast crack detection device that is connected to the computer 50 and the imaging information of the camera 30 is input to the computer 50,
The camera 30 and the bogie 20 are connected by a connecting device 31;
The connecting device 31 includes a connecting rod 32 having one end joined to the vehicle body 21 and protruding forward of the bogie 20, and a vertical rotating rod 33 which is a vertical rod whose upper end is connected to the tip end of the connecting rod 32, As a bar parallel to the connecting rod 32, the middle part of the horizontal rotating rod 34 is connected to the lower end of the vertical rotating rod 33, and as a rod orthogonal to the horizontal rotating rod 34 in plan view, the center of the horizontal rotating rod 34 is It is composed of a lateral extension rod 35 connected to the rear end and a counterweight 36 coupled to both ends of the lateral extension rod 35, respectively;
The front end of the connecting rod 32 and the upper end of the vertical rotating rod 33 are connected to be freely rotatable about a rotation axis parallel to the connecting rod 32 as an axis, and the lower end of the vertical rotating rod 33 and the middle of the horizontal rotating rod 34 are connected to the connecting rod (32) is connected so as to be freely rotatable about the axis of rotation perpendicular to the image information-based railway concrete ballast crack detection device, characterized in that the camera 30 is fastened to the front end of the horizontal rotating rod (34).
delete The method according to claim 1,
A sleeve 65 is joined to the vehicle body 21 of the bogie 20 so that the lifting rod 66, which is a vertical rod, can be freely raised and lowered;
A rotating system 62 mounted on one side of a driving wheel 61 running along a railroad track is joined to the lower end of the elevating bar 66;
A spring 67 is coupled to the elevating rod 66 between the sleeve 65 and the rotation system 62 , and the elastic force of the spring 67 acts between the sleeve 65 and the rotation system 62 , and the driving wheel 61 . always in close contact with this railroad track;
Image information-based railway concrete ballast crack detection device, characterized in that the display 63 of the rotation system 62 is included in the imaging area of the camera 30.
In the method for detecting cracks in railway concrete ballast (10) using the image information-based railway concrete ballast (10) crack detection device of any one of claims 1 and 3,
a photographing step (S10) in which the camera 30 mounted on the bogie 20 captures the map 10 so that the markers M1 and M2 formed on the map 10 are included;
an input step (S20) of the camera 30 inputting the raw information captured by the camera to the computer 50;
a storage step (S30) of the computer 50 receiving the imaged information and storing it in a storage device;
a positioning step (S41) in which the computer 50 retrieves the stored imaging information and binarizes the imaging information to extract the markers M1 and M2 having a preset shape;
a calculation step (S42) in which the computer 50 calculates the adjustment distance using the markers M1 and M2;
a correction step (S43) in which the computer 50 corrects the imaging information using the calculated adjustment distance to generate correction information;
Image information-based railway concrete ballast crack detection method, characterized in that the computer 50 binarizes the correction information, extracts the dark part of the crack (C), and stores the binarized correction information in a storage device (S50) .
The method according to claim 4, The first light emitter 41 and the second light emitter 42 are mounted on the bogie 20,
In the photographing step (S10), in the state in which the first light emitter 41 is turned on, the camera 30 captures the image 10 so that the markers M1 and M2 formed on the image 10 are included, and the first image information is generated. After the first light emitter 41 is turned off and the second light emitter 42 is turned on, the map 10 is photographed so that the markers M1 and M2 formed on the map 10 are included. is created;
In the input step (S20), the camera 30 inputs the first image information and the second image information, which are raw information captured by the camera, into the computer 50;
In the storage step (S30), the computer 50 receives the first image capturing information and the second capturing information and stores the received information in a storage device;
The first sensing information is applied as the sensing information, and the positioning step S41, the calculating step S42, and the correction step S43 are performed to generate the first correction information, and the second sensing information is applied as the sensing information. and the positioning step (S41), the calculation step (S42) and the correction step (S43) are performed to generate second correction information;
In the recording step (S50), the first and second correction information are respectively binarized to generate first binary information and second binary information, and logical values for each pixel of the first binary information and the second binary information are logically multiplied. Image information-based railway concrete ballast crack detection method, characterized in that by generating difference information by calculating in this way, the crack part (C), which is a dark part, is extracted and the difference information is stored in a storage device.

Images