JP6971731B2 - Inspection work support device, gap measurement method, and program - Google Patents

Description

The present invention relates to an inspection work support device for supporting inspection work of equipment such as a soundproof wall, a gap measurement method, and a program, and more particularly to a technique for measuring a gap between a support and a panel of a soundproof wall using a moving image.

If the gap between the pillars of the soundproof wall and the panel is widened and the pillars are no longer caught, the panel may fall. Therefore, it is important for train operation to inspect the parts where the panel may fall and keep the panel in a stable state. Currently, we are conducting visual inspections by walking patrols with a group of multiple people.

As a conventional technique using a photographing device or the like, there is a method of irradiating a slit light in a gap measurement using a photographed image (see Patent Document 1). There is also a method of photographing using a notch base (see Patent Document 2). There is also a method of inserting the measuring machine directly into the gap (see Patent Document 3).

Japanese Unexamined Patent Publication No. 2003-315020 Japanese Unexamined Patent Publication No. 2006-64597 Japanese Unexamined Patent Publication No. 2005-188993

Visual inspection by walking patrol takes time for inspection work (a total of about 26 man-days). Also, the inspection is not quantitative because it is visual.

An object of the present invention is to provide an inspection work support device, an inspection work support method, and a program for supporting inspection work of equipment such as a soundproof wall in order to solve the above problems.

The inspection work support device according to one aspect of the present invention is a device that supports the inspection work of the equipment, and is a shooting moving image registration unit for registering moving images obtained by simultaneously shooting the equipment of the upper and lower lines, and the shooting moving image registration. Gap measurement that measures the gap between the support column of the equipment and the panel based on the image of the support column extracted by the support column extraction unit and the support column extraction unit that extracts the image of the support column of the equipment from the moving image registered by the unit. The gist is that it has a department.

The gap measuring method according to one aspect of the present invention is a method in which an inspection work support device that supports inspection work of equipment measures a gap between a support column and a panel of the equipment, and is a moving image of the equipment of the upper and lower lines taken at the same time. The shooting moving image registration step for registering the image, the support column extraction step for extracting the image of the support column of the equipment from the moving image registered in the shooting moving image registration step, and the support column image extracted in the support column extraction step. The gist is to include a gap measurement step for measuring the gap between the support column and the panel of the equipment.

The gist of the program according to one aspect of the present invention is that it is a program for operating a computer as the inspection work support device.

According to the present invention, it is possible to provide an inspection work support device, an inspection work support method, and a program for supporting inspection work of equipment such as a soundproof wall.

It is explanatory drawing of the soundproof wall. It is explanatory drawing of the gap between a support and a panel at an end of a soundproof wall. It is explanatory drawing of the gap between a support and a panel in the middle part of a soundproof wall. It is explanatory drawing of the outline of this invention. It is explanatory drawing of the maintenance vehicle shown in FIG. It is a photograph of the name plate about the kilometer on the up line side. It is a photograph of the sign on the up line side other than the name plate about the kilometer on the up line side. It is a photograph of the name plate about the kilometer on the down line side. It is a photograph of the sign on the down line side other than the name plate about the kilometer on the down line side. It is a functional block diagram of the inspection work support device which concerns on 1st Embodiment. It is a flowchart which shows the operation of the inspection work support apparatus which concerns on 1st Embodiment. It is a flowchart which shows the operation of the inspection work support apparatus which concerns on 1st Embodiment. It is explanatory drawing of the kilometer name plate extraction process, the kilometer character recognition process, and the kilometer error correction process according to the first embodiment. It is explanatory drawing of the up line down line collision determination processing which concerns on 1st Embodiment. It is explanatory drawing of the down line kilometer interpolation adjustment processing which concerns on 1st Embodiment. It is a functional block diagram of the inspection work support device which concerns on 2nd Embodiment. It is a flowchart which shows the operation of the inspection work support apparatus which concerns on 2nd Embodiment. It is a flowchart which shows the operation of the inspection work support apparatus which concerns on 2nd Embodiment. It is explanatory drawing of the H steel column extraction process which concerns on 2nd Embodiment. It is explanatory drawing of the gap measurement processing which concerns on 2nd Embodiment.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the description of the drawings, the same or similar parts are designated by the same or similar reference numerals. However, it should be noted that the drawings are schematic, and the relationship between the thickness and the plane dimensions, the ratio of the thickness of each layer, etc. are different from the actual ones. Therefore, the specific thickness and dimensions should be determined in consideration of the following explanation. In addition, it goes without saying that parts having different dimensional relationships and ratios are included between the drawings.

<< First Embodiment >>
First, the first embodiment will be described. In the first embodiment, a technique for associating and managing distance marker information (position information in a railroad track) with a moving image will be described.

[Soundproof wall]
FIG. 1 is an explanatory diagram of the soundproof wall 100. The soundproof wall 100 is a semi-snow cover type soundproof wall 100 that has a soundproof effect, a snowproof effect, a windproof effect, and the like. Specifically, as shown in FIG. 1, the H-steel column (hereinafter, simply referred to as “support”) 110 is formed into a dogleg shape, which is expected to have an effect of preventing snow accumulation on the structure. A concrete panel is generally used for the soundproof wall panel (hereinafter, simply referred to as "panel") 120. The panel 120 is supported by a support 110 by a wedge 131.

As described above, when the gap between the support column 110 of the soundproof wall 100 and the panel 120 is widened and the support column 110 is no longer caught, the panel 120 is separated from the wedge 131 and floats in the air (circle in the figure). See mark). Depending on the state, the weight of the panel 120 cannot be supported, which may lead to the panel falling. Therefore, it is important for train operation to inspect the parts where the panel may fall and keep the panel in a stable state. Currently, we are conducting visual inspections by walking patrols with a group of multiple people.

[Gap between column and panel]
2A and 2B are explanatory views of a gap 141 between the support column 110 and the panel 120 at the end of the soundproof wall 100, FIG. 2A is a schematic view, and FIG. 2B is an actual photograph. Further, FIG. 3 is an explanatory view of a gap 141 between the support column 110 and the panel 120 in the intermediate portion of the soundproof wall 100, FIG. 3A is a schematic view, and FIG. 3B is an actual photograph. In these figures, the stanchion 110 exemplifies an H steel stanchion (steel). The H-steel column width W1 has a plurality of types such as 150 mm, 175 mm, and 200 mm depending on the height of the soundproof wall 100. In the figure, reference numeral W2 is the width of the center of the H-steel column, 131 is the wedge (rubber), 132 is the spacing material (rubber), 133 is the cushioning material (rubber), 134 is the end packing (rubber), 141 is the gap, 142. Indicates the amount of multiplication.

[Overview]
FIG. 4 is an explanatory diagram of an outline of the present invention. A kilometer-sized name plate (hereinafter, may be referred to as a "distance marker plate") that manages the position is installed on the track. As a method of installing a name plate of about a kilometer, there are a method of installing a pile on which a name plate is struck in a railroad track, a method of attaching it to a soundproof wall 100, and the like.

For example, as shown in FIG. 4, in the up line line 200A, the name plate 150A is installed every 10 m on the up line side kilometer (however, it is not installed at the points of 500 m and 1000 m). The up line is a train running line from the end point to the start point. On the other hand, in the down line line 200B, a name plate 150B about a kilometer on the down line side is installed every 500 m. The down line is a train running line from the starting point to the ending point.

FIG. 5 is an explanatory diagram of the maintenance vehicle 300 shown in FIG. In the present invention, the maintenance vehicle 300 travels on either the up line 200A or the down line 200B, and two cameras (soundproof wall photographing apparatus) 310 are installed toward the up line side and the down line side. Take pictures at the same time using. "Simultaneous" here does not mean complete time synchronization, but means "together". The traveling track and traveling direction of the maintenance vehicle 300 may be the opposite of this example.

The alternate long and short dash line in FIG. 5 indicates the shooting range of the camera 310. The lens (focal length) to be used, the arrangement, the number of lights, and the like of the camera 310 are appropriately selected according to the distance to the shooting target. The inspection work support device, which will be described later, automatically recognizes the kilometer name plate 150 imprinted on the captured moving image, adds track position information (kilometer information) to the moving image, and applies it to any frame. The position information in the track is determined.

[Sign]
FIG. 6 is a photograph of the up line side kilometer name plate 150A, and FIG. 7 is a photograph of signs on the up line side other than the up line side kilometer name plate 150A. The sign on the up line side is a yellow sign. As shown in FIG. 6, on the up line side kilometer name plate 150A, kilometers such as "380" and "276K300M" are displayed in horizontal black letters on a yellow background.

FIG. 8 is a photograph of the down line side kilometer name plate 150B, and FIG. 9 is a photograph of signs on the down line side other than the down line side kilometer name plate 150B. The sign on the down line side is a white sign. As shown in FIG. 8, on the down line side kilometer name plate 150B, kilometers such as "278" and "276 1/2" are displayed in vertical black characters on a white background.

[Configuration of inspection work support device]
FIG. 10 is a functional block diagram of the inspection work support device 10 according to the first embodiment. This inspection work support device 10 is a device that supports inspection work of equipment in a railroad track such as a soundproof wall 100, and is functionally a shooting moving image registration unit 11, an image processing parameter registration unit 12, and a kilometer or so. The name plate extraction unit 13, the kilometer nameplate correction unit 14, the kilometer character recognition (OCR) unit 15, the recording unit 16, the kilometer error correction unit 17, the upline downline matching unit 18, and the moving image kilometer. The information DB unit 19 is provided. The up line down line matching unit 18 includes an up line down line collision determination unit 18-1 and a down line kilometer interpolation adjustment unit 18-2.

The photographing moving image registration unit 11 registers the moving image photographed by the soundproof wall photographing apparatus 310. Specifically, the running line section (0: up line / 1: down line), running direction (0: starting point ⇒ ending point / 1: ending point ⇒ starting point), starting point station name, ending point station name at the time of shooting , Shooting soundproof wall (0: up line side / 1: down line side), start kilometer, end kilometer, image resolution, left end of image (0: start point / 1: end point), right edge of image (0: Information on the start point / 1: end point) and the rotation display (-1: left 90 degrees / 0: 0 degrees / 1: right 90 degrees) is registered in the shooting moving image.

The image processing parameter registration unit 12 registers the image processing parameters. Specifically, the parameters of image processing (size, color, etc. of the kilometer nameplate) required for the kilometer nameplate extraction process and the kilometer character recognition process are registered.

The kilometer nameplate extraction unit 13 performs a kilometer nameplate extraction process. Specifically, based on the image processing parameters registered by the image processing parameter registration unit 12, an image of a kilometer name plate is extracted from the moving image registered by the shooting moving image registration unit 11. In addition, it is determined whether or not the image of the extracted kilometer name plate matches the size of the target name plate.

The kilometer nameplate correction unit 14 performs the kilometer nameplate correction process. Specifically, the tilt correction is performed on the image of the kilometer nameplate 150A on the upstream side extracted by the kilometer nameplate extraction unit 13. On the other hand, the image of the down line side kilometer nameplate 150B extracted by the kilometer nameplate extraction unit 13 is subjected to image correction of left-right division / projective projection conversion.

The kilometer character recognition unit 15 performs kilometer character recognition processing. Specifically, the image corrected by the name plate correction unit 14 about kilometer is subjected to character recognition (OCR) of the numerical value of "0-9", the character of "K, M", and the character of "-" symbol, and kilogram. Determine the degree.

The kilometer error correction unit 17 performs a kilometer error correction process. Specifically, whether or not the value of the kilometer determined by the kilometer character recognition unit 15 is correct is corrected by considering the traveling direction, the traveling speed, the distance between the nameplates of the kilometer, and the like.

The recording unit 16 is a recording device that records various data.

The up line down line matching unit 18 performs an up line down line matching process. Specifically, the up line down line collision determination unit 18-1 compares the result (kilo) of the down line kilometer name plate extraction process / kilometer character recognition process with the kilometer estimated on the upline side. .. If the matching results do not match, it is considered that the kilometer estimated on the up line side is correct, the search is performed again around it, and the frame synchronization when the up line side and the down line side are shot at the same time. Determine the deviation (frame deviation). Further, the down line kilometer interpolation adjustment unit 18-2 interpolates the down line kilometer information from the up line side kilometer information using the result determined by the up line down line collision determination unit 18-1. Interpolate and adjust the down line kilometer. Based on the frame value of the synchronization deviation, the information of about 10 m unit detected on the up line side is interpolated and inserted on the down line side.

The moving image kilometer information DB unit 19 manages the moving image by associating the kilometer information with the moving image. Specifically, the number of frames and the number of kilometers and meters are registered in the DB for each of the up line and the down line.

[Operation of inspection work support device]
FIG. 11 is a flowchart showing the operation of the inspection work support device 10 according to the first embodiment. Hereinafter, the operations of the kilometer name plate extraction unit 13, the kilometer nameplate correction unit 14, and the kilometer character recognition unit 15 will be described with reference to FIG.

First, the moving image → still image conversion is performed, and then the RGB → HSI conversion is performed (step S1 → S2). The moving image → still image conversion is to convert a moving image taken by the soundproof wall photographing apparatus 310 into a still image. The RGB → HSI conversion is to convert from the RGB color space to an image in the HSI color space in order to easily identify the characteristic color (upline side: yellow, downline side: white) of the name plate.

In addition to HSI conversion, methods such as HSV conversion and HSL conversion can also be used. That is, HSI is an HSI color space of a color system described by hue (H: hue), saturation (S: saturation), and lightness (I: intensity), and is HSV (V: value) or HSL (L). : lightness). This is because it is used almost the same as intensity = value = Lightness.

Next, an image of a kilometer name plate is extracted from the HSI-converted image, and it is determined whether or not there is a designated color region (step S3 → S4). In the case of up line side photography, it is determined whether or not there is a yellow area, and in the case of down line side photography, it is determined whether or not there is a white area.

Here, if there is a designated color area, it is determined whether or not the size of the designated color area is correct (step S5). If the size of the designated color area is correct, the extracted image of the kilometer name plate is corrected, character recognition of the kilometer is performed, and it is determined whether or not the designated character symbol is recognized (step S6 → S7 → S8). In the case of up line side shooting, it is determined whether the numerical value of "0-9" and the character "K, M" are recognized, and in the case of down line side shooting, the numerical value of "0-9", "-" "Determine if the symbol is recognized.

Here, if the designated character symbol is recognized, internal output is performed for about a kilometer (step S9). That is, the same kilometer name plate is extracted from a plurality of frames, and the kilometer is obtained by OCR. The internal output of about a kilometer is the internal output of the data.

The following is an example of the internal output of the kilometer on the up line side. From left to right, the file name, number of frames, distance from the center of the image, name plate type, and OCR result are shown. Since the distance here is the distance on the image, the unit is pixel. The center of the image is set to 0, and a "-" sign is attached to the left side of the center of the image and a "+" sign is attached to the right side.

Up line side_C0003, 1845, -107.50, yellow name plate, 130
Up line side_C0003, 1846, -62.50, yellow name plate, 130
Up line side_C0003, 1847, -16.00, yellow name plate, 130
Up line side_C0003, 1848, 30.00, yellow name plate, 130
Up line side_C0003, 1849, 77.22, yellow name plate, 130
Up line side_C0003, 1850, 120.76, yellow name plate, 130
Next, it is determined whether or not the kilometer is the same, and if it is not the same kilometer, the previous kilometer is output as a temporary DB (step S10 → S11). "Outputting the previous kilometer to the temporary DB" means outputting the kilometer result of the frame closest to the center of the image to the temporary DB among the "results of the internal output of the kilometer". In the case of this example of the up line side, "up line side_C0003, 1847, -16.00, yellow name plate, 130" is output as a temporary DB (step S11).

Similarly, an example of the internal output of about a kilometer on the down line side is shown. In the case of this example of the downlink side, "downline side_C0002, 2545, 30.21, white name plate (left side), 280" is output as a temporary DB (step S10 → S11).

Down line side_C0002, 2544, 191.42, white name plate (left side), 280
Down line side_C0002, 2544, 191.42, white name plate (right side), 280
Down line side_C0002, 2545, 30.21, white name plate (left side), 280
Down line side_C0002, 2545, 30.21, white name plate (right side), 280
Down line side_C0002, 2546, -132.00, white name plate (left side), 280
Down line side_C0002, 2546, -132.00, white name plate (right side), 280
The frame is moved to the next frame, and the same process is repeated (step S12 → S1 → ...). When the final frame is reached, the process proceeds to the next step (step S12 → S13). In the case of up line side shooting, the process moves to error correction processing for about a kilometer, and in the case of down line side shooting, it moves to up line down line collision determination / interpolation adjustment processing.

FIG. 12 is a flowchart showing the operation of the inspection work support device 10 according to the first embodiment. Hereinafter, the operation of the error correction section 17 and the upline / downline matching section 18 will be described with reference to FIG.

First, the kilometer error correction unit 17 corrects the error about the kilometer information on the upstream line output in the temporary DB (steps S21 → S22). Then, the kilometer information on the up line side after the error correction is output to the DB (step S24), and the estimated 500 m, 1 km kilometer of the kilometer information on the up line side after the error correction is performed. The information frame is calculated (step S23).

On the other hand, the up line down line matching unit 18 checks the kilometer information by narrowing down the name plate extraction range for the kilometer information on the down line side output in the temporary DB (steps S26 → S27). If the name plate has already been extracted or the OCR is successful, the frame difference between the estimated value on the up line side and the extracted value on the down line side is calculated (step S28 → S29). Then, in consideration of this frame difference, interpolation interpolation is performed by interpolating about every 10 m and every 100 m on the up line side to about the kilometer on the down line side (steps S25 and S33). At this time, the synchronization shift of the frame when the up line side and the down line side are photographed at the same time is adjusted in a section of about 500 m and 1 km (step S33). Finally, the kilometer information on the down line side after being adjusted in this way is output to the DB (step S34).

If NO is determined in step S28, reprocessing is performed (step S30). Reprocessing is to perform the kilometer name plate extraction process and the kilometer character recognition process again in a limited number of frames. If the reprocessing is successful (step S31: YES), the process proceeds to step S29, and if the reprocessing fails (step S31: NO), it is calculated from the estimated value on the uplink side and the frame difference before and after (step S32). ), Move to step S29.

[Kilo name plate extraction process, Kilo character recognition process]
FIG. 13 is an explanatory diagram of a kilometer name plate extraction process, a kilometer character recognition process, and a kilometer error correction process according to the first embodiment. On the horizontal axis, the frame position (assumed) where the name plate is thought to be about a kilometer is indicated by the tick mark "|".

Hereinafter, the case of the up-line side photographed moving image will be described with reference to FIG. 13 (a).

First, the name plate extraction process (process 1) is performed for about kilograms. That is, the frame in the yellow region (threshold value is set by the size of the image width / height) by RGB → HSI conversion is extracted. The part where the yellow area can be extracted is indicated by "○". On the soundproof wall 100, a sign in a yellow area is installed in addition to the name plate of about a kilometer, but since it is smaller than the setting threshold value of the image size, it is excluded by the size determination. However, it is not possible to extract the missing part of the name plate due to peeling.

Next, a kilometer character recognition process (process 2-1) and a kilometer error correction process (process 2-2) are performed. That is, the number 0-9 and the letters K and M are recognized for the extraction location (name plate). The part where character recognition was successful is indicated by "○", and the part where character recognition failed is indicated by "×". However, here, the part of incorrect character recognition is also "○". Temporarily retains the kilometer value after character recognition, corrects errors, and corrects to the correct kilometer value. Furthermore, for objects that could not recognize characters, if the frame value estimated from the number of frames per 10 m matches the extracted location from the location where character recognition was successful, a kilometer value is given to that frame. Process and finalize the correspondence between the frame and the kilometer.

Next, a case of a moving image taken on the down line side will be described with reference to FIG. 13 (b).

First, the name plate extraction process (process 3) is performed for about kilograms. That is, a frame in a white region (threshold value is set by the size of the image width / height) by RGB → HSI conversion is extracted. The part where the white area can be extracted is indicated by "○". On the soundproof wall 100, a sign in a white area is installed in addition to the name plate of about a kilometer, but if it is smaller than the setting threshold value of the image size, it is excluded by the size determination.

Next, a character recognition process (process 4) for about a kilometer is performed. That is, the character recognition of the number 0-9 and the-symbol is performed for the extraction point (name plate). The part where character recognition was successful is indicated by "○", and the part where character recognition failed is indicated by "×". However, here, the part of incorrect character recognition is also "○". Therefore, as will be described later, the estimation frame extracted from the up line and the extraction frame of the down line are matched, and the interpolation adjustment process is performed.

[Up line / Down line collision judgment processing]
By the way, even if the shutter is pressed at the same time in the soundproof wall photographing apparatus 310 and shooting is started, the time cannot be synchronized in frame units. Therefore, the up line side shooting moving image and the down line side shooting moving image are matched and adjusted in frame units. In order to perform perfect time synchronization in frame units, special equipment such as a device that controls an external trigger and an external device input on the camera side is required, but in the present invention, countermeasures are taken by image processing. ..

FIG. 14 is an explanatory diagram of an up line / down line collision determination process according to the first embodiment. The horizontal axis shows the frame numbers of 276K, 276K500M, 277K, 277K500M, ... Estimated from the up-line side photographed moving image.

As shown in FIG. 14, the difference (usually within ± 0.5 seconds) within ± 1 second (usually within ± 0.5 seconds) by comparing the frame number at the kilometer position estimated from the up line with the result of the name plate extraction / OCR of the down line (within ± 0.5 seconds). If it is ± 30 frames in terms of frames), it is determined that the name plate can be extracted on the down line. Furthermore, if the OCR result is about an estimated kilometer, the accuracy will be improved. Even if the name plate has been extracted but OCR has not been performed, it is used as an estimated value (the part in bold "277K500M"). On the other hand, the part where the deviation of ± 30 frames occurs is considered to have been extracted from the name plate other than about kilograms, so that part is excluded (the part of "277K500M" in fine print).

[Downline kilometer interpolation adjustment processing]
FIG. 15 is an explanatory diagram of the downlink kilometer interpolation adjustment process according to the first embodiment. As shown in this figure, after the up line down line collision determination process, the down line kilometer interpolation adjustment process is performed. Specifically, in consideration of the frame difference between the up line and the down line, linear interpolation interpolation is performed for the kilometer information in units of 10 m on the up line side to the down line side. The white arrows in the figure are arrows for showing the state before and after the interpolation, and "..." means the interpolated information of about 10 m unit.

[summary]
As described above, the inspection work support device 10 according to the first embodiment is a device that supports the inspection work of the soundproof wall 100, and is a photographing device for registering a moving image obtained by simultaneously photographing the soundproof wall 100 of the upper and lower lines. The moving image registration unit 11, the kilometer name plate extraction unit 13 that extracts the image of the kilometer name plate from the moving image registered by the shooting video registration unit 11, and the kilometer nameplate extracted by the kilometer nameplate extraction unit 13. The kilometer character recognition unit 15 that recognizes characters on the image and determines the kilometer information, and the kilometer information determined by the kilometer character recognition unit 15 on the moving image registered by the shooting moving image registration unit 11. It is provided with a moving image kilometer information DB unit 19 that is linked and managed. As a result, the distance marker information can be associated with the moving image of the soundproof wall 100 of the upper and lower lines taken at the same time, so that the inspection work of the soundproof wall 100 can be efficiently performed using this moving image.

The method of using the moving image is not particularly limited, but it can be used, for example, in the gap measuring method described in the second embodiment (described later). Of course, the moving image may be reproduced by a computer, and the operator may visually inspect the equipment in the track based on the moving image. Since the distance marker information is associated with the moving image, if a defect is found, the position in the track can be easily identified.

Further, an up line / down line matching unit 18 may be provided to ensure consistency between the kilo information on the up line side and the kilo information on the down line side recognized by the character recognition unit 15. As a result, it is possible to accurately obtain information on the kilometer on the up line side and kilometer information on the down line side.

Specifically, even if the up line down line matching unit 18 interpolates the kilometer information on the down line side from the kilometer information on the up line side and adjusts the kilometer information on the down line side. good. As a result, based on the frame value of the synchronization deviation, the information of about 10 m unit detected on the up line side can be interpolated and inserted on the down line side.

Further, the kilometer nameplate correction unit 14 is provided to perform image correction on the image of the kilometer nameplate on the up line side and the image of the kilometer nameplate on the downline side extracted by the kilometer nameplate extraction unit 13 by different methods. May be good. As a result, it is possible to perform image correction according to the characteristics of the kilometer name plate on the up line side and the kilometer name plate on the down line side.

Further, if there is an error in the kilometer information determined by the kilometer character recognition unit 15, the kilometer error correction unit 17 may be provided to correct the error portion. As a result, the error portion is corrected, and it is possible to obtain accurate information about the kilometer.

The present invention can be realized not only as an inspection work support device 10, but also as a distance marker information management method using a characteristic processing unit included in the inspection work support device 10 as a step, or an inspection work support device. It can also be realized as a program for operating the computer as 10. Needless to say, such a program can be distributed via a recording medium such as a CD-ROM or a transmission medium such as the Internet.

<< Second Embodiment >>
Next, only the differences between the second embodiment and the first embodiment will be described. In the second embodiment, a technique for accurately measuring the gap 141 between the support column 110 of the soundproof wall 100 and the panel 120 using a moving image will be described. In the following explanation, it is assumed that the distance marker information is associated with the moving image. The method of associating the distance marker information with the moving image may be the distance marker information management method described in the first embodiment, or may be another method.

[Configuration of inspection work support device]
FIG. 16 is a functional block diagram of the inspection work support device 20 according to the second embodiment. The inspection work support device 20 is a device that supports inspection work of equipment in the track such as a soundproof wall 100, and is functionally a shooting moving image registration unit 21, an image processing parameter registration unit 22, and a moving image. The kilometer information DB unit 23, the H-beam extraction unit 24, the recording unit 25, the gap measurement unit 26, the gap measurement result output unit 27, the risk determination unit 28, and the gap measurement ledger output unit 29. Be prepared.

The photographing moving image registration unit 21 registers the moving image photographed by the soundproof wall photographing apparatus 310. Specifically, the running line section (0: up line / 1: down line), running direction (0: starting point ⇒ ending point / 1: ending point ⇒ starting point), starting point station name, ending point station name at the time of shooting , Shooting soundproof wall (0: up line side / 1: down line side), start kilometer, end kilometer, image resolution, left end of image (0: start point / 1: end point), right edge of image (0: Information on the start point / 1: end point) and the rotation display (-1: left 90 degrees / 0: 0 degrees / 1: right 90 degrees) is registered in the shooting moving image.

The image processing parameter registration unit 22 registers the image processing parameters. Specifically, the image processing parameters (binarization threshold value, detection score threshold value, column center width, center binarization threshold value) required for H steel column extraction processing and gap measurement processing, and column template image (uplink). Register the line side, down line side, end / intermediate part, etc.) and H steel column width (mm).

The moving image kilometer information DB unit 23 corresponds to the moving image kilometer information DB unit 19 in the first embodiment. That is, information about a kilometer is associated with a moving image for management. Specifically, the number of frames and the number of kilometers and meters are registered in the DB for each of the up line and the down line.

The H-steel column extraction unit 24 performs an H-steel column extraction process. Specifically, the image of the H steel column is extracted according to the image processing parameters (binarization threshold value, detection score threshold value, column center width) registered by the image processing parameter registration unit 22 and the column template image.

The recording unit 25 is a recording device that records various data.

The gap measurement unit 26 performs the gap measurement process. Specifically, the image processing parameters (width of the center of the column, binarization threshold of the center) registered by the image processing parameter registration unit 22 based on the image of the H steel column extracted by the H steel column extraction unit 24. ), The gap 141 between the support column 110 and the panel 120 is measured.

The gap measurement result output unit 27 performs gap measurement result output processing. Specifically, the amount of hooking is calculated from the gap measured by the gap measuring unit 26, the degree of danger is determined from the calculated amount of hooking, and the result is output in units of H steel columns. For example, the hooking amount 142 can be calculated by (H steel column width W1-H steel column center width W2) ÷ 2-gap 141 (see FIG. 2).

The risk level determination unit 28 performs a risk level determination process. Specifically, the degree of danger (construction priority) is determined based on the gap (hanging amount) measured by the gap measuring unit 26.

The gap measurement ledger output unit 29 performs the gap measurement ledger output process. Specifically, using the results output by the gap measurement result output unit 27 and the risk level determination unit 28, a form is used as a ledger for the gap, the amount of hooking, the degree of risk, etc. of the soundproof wall panel unit (posts at the left and right ends). Output.

[Operation of inspection work support device]
FIG. 17 is a flowchart showing the operation of the inspection work support device 20 according to the second embodiment. Here, the operations of the H-steel column extraction unit 24, the gap measurement unit 26, and the gap measurement result output unit 27 will be described.

First, moving image → still image conversion is performed, RGB → gray scale conversion is performed, and a gray scale image is binarized (step S41 → S42 → S43). The binarization process is a process for facilitating the distinction between H-steel columns, concrete panels, gaps, and the like.

Next, H-steel column template matching is performed (step S44). H-steel column template matching is to perform image matching using the H-steel column template image. Specific examples will be described in detail later.

Next, if the detection score is larger than the threshold value as a result of the H steel column template matching, the extracted H steel column is determined (step S45 → S46). In the determination of the extracted H steel column in step S46, it is desirable to perform the following processes (1) and (2).

(1) Extraction of columns in consideration of the interval between columns (1.5m, 2m, 2.5m, 3m) The panel width of the installed soundproof wall 100 (interval between columns 110) is calculated from the extracted column frame numbers. ), The frame position of the support column 110 that appears next is estimated. In addition, the difference between the kilometer of the extracted support column 110 and the kilometer immediately before is calculated, and extraction errors (when exceeding 3 m, less than 1 m, etc.) are excluded based on the difference.

(2) Consideration of strut position height The strut position height changes to some extent due to the influence of Kant, but it is almost constant. Therefore, taking into consideration that the height of the extracted column positions is almost constant, the extraction error is eliminated.

Next, when the determination condition of the extracted H steel column in step S46 is satisfied, an image of the extracted H steel column is output, the central position of the center of the column is specified, and the binarization process is performed based on the binarization threshold value. Then, it is determined whether or not the width of the center of the column of the image is the same as the width of the center of the column (step S47 → S48 → S49 → S50). Here, the determination conditions in step S46 are (1) that the column spacings match 1.5 m, 2.0 m, 2.5 m, and 3 m, and (2) the extracted H-steel column position is at the left end. It is a series of sets of the middle part and the right end part, and there is no extreme change in the numerical value of the height on the image. That is, (1) in the case of a column spacing of 1 m or less or 4 m or more, or (2) the extracted H-steel column position is a series of sets of the left end part, the middle part, and the right end part, and the numerical value of the height on the image is extreme. If there is a change, it is determined as NO in step S46. If the width of the center of the column in the image is not the same as the width of the center of the column, the binarization threshold is shifted up and down and the search is performed again (step S50 → S51).

Specifically, the value of the width of the center of the end of the column is set as the specified value, one line is acquired with the width of 1 pixel in the horizontal direction, and the binarization is performed at the median value of the maximum and minimum brightness. When a peak with brightness close to the number is found, that position is set as the center of the column, and when no peak is found, the binarization threshold is shifted up and down, and the binarization process is performed again. As a result, if the width of the center of the column becomes thinner or thicker when binarized due to the influence of light or the like, the influence can be avoided.

On the other hand, when the width is the same as the predetermined width of the center of the column, each of the concrete panel, the gap, the column, the gap, and the concrete panel is distinguished from the brightness value (step S50 → S52). Specifically, the central part of the column end is extracted, the low-brightness part that spreads on both sides of the extracted central part is searched, and the part where the brightness is increased is recognized as the panel 120, and the panel is displayed from the column end. The gap is up to 120. Thereby, when the support column 110 is an H steel support column, the H steel support column, the panel 120, and the gap 141 can be accurately distinguished.

Next, the gap is measured from the distinction position (step S53). At this time, the resolution of the image is set and the actual size is obtained from the pixels. Further, the amount of hooking is calculated from the gap based on the width of the H steel column, and the gap measurement result of the extraction column is output (step S54 → S55).

The frame is moved to the next frame, and the same processing is repeated (step S56 → S41 → ...). When the final frame is reached, the process proceeds to the next step (step S57).

FIG. 18 is a flowchart showing the operation of the inspection work support device 20 according to the second embodiment. Here, the operation of the risk level determination unit 28 and the gap measurement ledger output unit 29 will be described.

First, the left and right end columns (starting point column and ending point column) of the soundproof wall panel are specified based on the extraction column clearance measurement result (step S61 → S62). Specifically, taking into account the information of the application template (end (left / right), middle part) at the time of template matching and the number of struts, the set of start point struts and end struts (left end-middle part, middle part) -Middle part, middle part-right end part) is decided.

Next, the degree of danger is determined and the ledger is output (step S63 → S64). In the risk level determination, a plurality of levels of risk level (construction priority) are determined according to the amount of hooking of the left and right end columns, the numerical value of the gap, and the panel position (the number of the panel from the end). For example, it may be determined that there are five levels of risk and no problem (not subject to construction).

After that, the same process is repeated until the final panel (step S65 → S66 → S61).

[H steel column extraction process]
FIG. 19 is an explanatory diagram of the H-steel column extraction process according to the second embodiment. FIG. 19A shows an example of the left end template image 31, the middle template image 32, and the right end template image 33. FIG. 19B shows how H-steel column template matching is performed using these template images 31 to 33.

The support column 110 of the soundproof wall 100 is configured in the order of left end portion → middle portion → right end portion (or vice versa). Therefore, the left end is followed by the middle part (rarely the right end), the middle part is followed by the middle or right end, and the right end is followed by the left end support with a short interval. In consideration of this, the applicable template is selected from the template images 31 to 33 in a timely manner.

For example, as shown in FIG. 19B, the left end template image 31 is initially selected, and the left end template image 31 extracts the image 41 of the H steel column. Next, the intermediate portion template image 32 or the right end portion template image 33 is selected, and the images 42, 43, 44 of the H steel column are sequentially extracted by the intermediate portion template image 32. The next image 45 is excluded as an extraction error because the extraction height is not within the predetermined range. Next, the intermediate portion template image 32 or the right end portion template image 33 is selected, and the image 46 of the H steel column is extracted by the intermediate portion template image 32. Finally, the intermediate portion template image 32 or the right end portion template image 33 is selected, and the image 47 of the H steel column is extracted by the right end portion template image 33.

[Gap measurement processing]
FIG. 20 is an explanatory diagram of the gap measurement process according to the second embodiment. Hereinafter, the gap measurement process will be described in more detail with reference to FIG.

As shown in this figure, it was decided to extract the central part of the end of the column and search for the low-brightness part spreading on both sides from that position. That is, first, the pixel of the width of the center of the end of the column is measured and set as a specified value. One line is acquired with a width of 1 pixel in the horizontal direction, binarization is performed at the median of the maximum and minimum brightness, and if there is a peak of brightness close to the specified number of pixels, that is used as the center of the column. If no peak is found within the specified width, the binarization threshold is shifted up and down from the median, and the binarization process is performed again. When the central part of the column can be extracted, the parts with low brightness are measured on both sides, the part where the brightness is increased is recognized as a concrete panel, and the number of pixels is measured with the concrete panel as a gap.

[summary]
As described above, the inspection work support device 20 according to the second embodiment is a device that supports the inspection work of the soundproof wall 100, and is a photographing device for registering a moving image obtained by simultaneously photographing the soundproof wall 100 of the upper and lower lines. The moving image registration unit 21, the H steel support column extraction unit 24 that extracts the image of the support column 110 of the soundproof wall 100 from the moving image registered by the shooting moving image registration unit 21, and the support column extracted by the H steel support column extraction unit 24. Based on the image of 110, the support column 110 of the soundproof wall 100 and the gap measuring unit 26 for measuring the gap 141 of the panel 120 are provided. As a result, the gap 141 between the support column 110 of the soundproof wall 100 and the panel 120 can be accurately measured using the moving image, so that the inspection work of the soundproof wall 100 can be efficiently performed.

It is desirable that the distance marker information is associated with the moving image. The method of associating the distance marker information with the moving image may be the distance marker information management method described in the first embodiment, or may be another method.

Further, the gap measuring unit 26 extracts the central portion of the column end, searches for low-brightness portions spreading on both sides of the extracted central portion, recognizes the increased brightness as the panel 120, and recognizes the column end. The gap from the portion to the panel 120 may be used as a gap. Thereby, when the support column 110 is an H steel support column, the H steel support column, the panel 120, and the gap 141 can be accurately distinguished.

Further, the gap measuring unit 26 sets the pixel of the width of the central portion of the column end as a specified value, acquires one line with a width of 1 pixel in the horizontal direction, binarizes it at the median of the maximum and minimum brightness, and defines it. When a peak of brightness close to the number of peaks of the value is found, that position may be set as the center of the column, and when no peak is found, the binarization threshold value may be shifted up or down from the median. As a result, even if the width of the center of the column is thinned or thickened by the binarization process due to the influence of light or the like, there is an effect that the influence is less likely to be affected.

Further, the H-steel column extraction unit 24 has template images 31 to 33 of the columns at the left end portion, the intermediate portion, and the right end portion, and the left end portion, the intermediate portion, and the right end portion are in this order, or the right end portion, the intermediate portion, and the left end portion. Template matching may be performed in order. Thereby, the application template can be selected in a timely manner according to the configuration of the soundproof wall 100.

Further, a risk level determination unit 28 may be provided, which calculates the amount of hooking from the gap 141 measured by the gap measuring unit 26 and determines the degree of danger based on the calculated amount of hooking. This makes it possible to easily determine the timing and necessity of construction based on the degree of danger.

The present invention can be realized not only as an inspection work support device 20, but also as a gap measurement method using a characteristic processing unit provided in the inspection work support device 20 as a step, or as an inspection work support device 20. It can also be realized as a program for operating a computer. Needless to say, such a program can be distributed via a recording medium such as a CD-ROM or a transmission medium such as the Internet.

Further, in the above description, the semi-snow-covered soundproof wall 100 is exemplified as the equipment inside the track, but the equipment inside the track is not limited to the soundproof wall 100. For example, the present invention can be applied even in a place where there is no soundproof wall 100, such as in a tunnel, as long as a kilometer name plate is installed in the tunnel (equipment in the track).

Further, in the above description, the in-line equipment is mentioned as an example of the equipment, but the scope of application of the present invention is not limited to the in-line equipment. For example, the present invention can be applied if a distance marker (equipment) is installed on a highway or a river.

<< Other embodiments >>
As mentioned above, some embodiments have been described, but the statements and drawings that form part of the disclosure are exemplary and should not be understood as limiting each embodiment. This disclosure will reveal to those skilled in the art various alternative embodiments, examples and operational techniques.

As such, each embodiment includes various aspects not described herein.

10 ... Inspection work support device 11 ... Shooting video image registration unit 12 ... Image processing parameter registration unit 13 ... Kilo name plate extraction unit 14 ... Kilo name plate correction unit 15 ... Kilo character recognition unit 16 ... Recording unit 17 ... Kilo error Correction unit 18 ... Upline downline matching unit 18-1 ... Upline downline collation determination unit 18-2 ... Downline kilometer interpolation adjustment unit 19 ... Moving image kilometer information DB unit 20 ... Inspection work support device 21 ... Shooting Video registration unit 22 ... Image processing parameter registration unit 23 ... Video processing kilometer information DB unit 24 ... H steel column extraction unit 25 ... Recording unit 26 ... Gap measurement unit 27 ... Gap measurement result output unit 28 ... Danger level determination unit 29 … Gap measurement ledger output unit 100… Soundproof wall (equipment in the line)
110 ... H steel support (support)
120 ... Soundproof wall panel (panel)
141 ... Gap 150 ... Kilometer name plate (distance marker plate)
150A ... Up line side kilometer name plate 150B ... Down line side kilometer name plate 200A ... Up line line 200B ... Down line line 300 ... Maintenance car 310 ... Soundproof wall photographing device

Claims (7)

It is an inspection work support device that supports the inspection work of equipment equipped with panels fitted between multiple H-steel columns.
A shooting moving image registration unit that registers moving images taken at the same time with the equipment of the upper and lower lines,
A column extraction unit that extracts an image of the H-steel column from a moving image registered by the photographed moving image registration unit, and a column extraction unit.
Based on the image of the H-steel column extracted by the column extraction unit, a gap measuring unit for measuring the gap between the H-steel column and the panel is provided.
The gap measuring unit extracts the central portion of the column end portion of the H-steel column, searches for low-luminance portions spreading on both sides of the extracted central portion, and recognizes the increased brightness as the panel. , An inspection work support device characterized in that the gap from the end of the support column to the panel is the gap. The gap measuring unit takes a pinxel of the width of the center of the end of the column as a specified value, acquires one line with a width of 1 pinxel in the horizontal direction, binarizes it at the center value of the maximum and minimum brightness, and performs the above-mentioned specified value. claim when luminance peak close to pixel number value is found that position the center portion of the H steel struts, when the peak is not found, characterized in that shifting the binarization threshold vertically 1 Inspection work support device described in. The support column extraction unit has a template image of the H-steel column at the left end portion, the middle portion, and the right end portion, and template matching is performed in the order of the left end portion, the middle portion, and the right end portion, or in the order of the right end portion, the middle portion, and the left end portion. The inspection work support device according to any one of claims 1 to 2. Further, the amount of hooking is calculated from the gap measured by the gap measuring unit using (Formula 1).
Amount of engagement = (width of H steel column-width of center of H steel column) ÷ 2-gap (Formula 1)
Inspection support apparatus according to any one of claims 1 to 3, further comprising a risk determination unit for determining the risk based on the consuming amount the calculated. The inspection work support device according to any one of claims 1 to 4 , wherein the equipment is a semi-snow-covered soundproof wall. An inspection work support device that supports inspection work of equipment provided with a panel fitted between a plurality of H-steel columns is a gap measurement method for measuring the gap between the H-steel columns and the panel.
The shooting moving image registration step to register the moving image of the equipment of the upper and lower lines taken at the same time,
A column extraction step for extracting an image of the H-steel column from a moving image registered in the photographed moving image registration step, and a column extraction step.
Based on the image of the H steel strut extracted by the post extraction step, seen including a gap measuring step of measuring the gap between the steel H strut and the panel,
In the gap measurement step, the central portion of the column end portion of the H-steel column is extracted, the low-luminance portion spreading on both sides of the extracted central portion is searched, and the portion where the brightness is increased is recognized as the panel. , A gap measuring method, characterized in that the gap is set from the end of the support column to the panel. A program for operating a computer as an inspection work support device according to any one of claims 1 to 5.

Images