KR20230066994A - Apparatus and method for monitoring cracks on surfaces covered with self-healing repair mortars using image processing techniques - Google Patents

Abstract

The monitoring device of the present invention may include: a photographing unit that acquires a photographed image of a concrete surface; and an analysis unit that detects and analyzes cracks through the analysis of the photographed image. Therefore, it is possible to monitor cracks on surfaces covered with self-healing repair mortars through image analysis.

Description

Apparatus and method for monitoring cracks on surfaces covered with self-healing repair mortars using image processing techniques}

The present invention relates to an apparatus and method for monitoring cracks in a target surface on which self-healing repair mortar is applied.

Recently, with the aging of social infrastructure in Korea, interest in inspection and maintenance of facilities and related research are rapidly increasing.

In performing safety diagnosis and inspection of facilities, the most basic investigation item is exterior condition investigation.

Currently, most of the exterior inspection methods for concrete cracks rely on visual inspection by manpower, which makes it difficult to accurately quantify and lacks objectivity. In addition, although the investigated cracks are recorded on the exterior survey network map, they remain at the level of approximate location marks rather than the shape and size of actual cracks, making efficient data management impossible. Accordingly, the need for efficient facility inspection and survey technology that reduces time and cost by minimizing the input of manpower is emerging.

Korean Registered Patent Publication No. 2215035 discloses a technology that easily monitors cracks in concrete structures as well as self-healing of self-healing repair mortar from cracks by combining IoT IoT technology and wireless communication service technology from a remote location rather than on-site. It is becoming.

Korean Registered Patent Publication No. 2215035

The present invention is to provide a monitoring device and method for monitoring cracks on the construction surface through image analysis of the self-healing repair mortar construction surface.

The monitoring device of the present invention includes a photographing unit for obtaining a photographed image of a concrete surface; An analysis unit that detects and analyzes cracks through analysis of the photographed image; may include.

The monitoring method of the present invention includes a setting step of setting a region of interest (ROI) in a photographed image of a concrete surface; a binarization step of converting the ROI into a binarized image; an extraction step of applying a morphology technique to the binarized image so that a first candidate region of a crack having continuity and blobs removed is detected; determining a second candidate region by ANDing an image including the first candidate region with the binarized image; and an analysis step of tracking cracks in the second candidate region determined in the determining step and calculating a width of the tracked crack.

The monitoring device of the present invention can detect cracks using a photographed image of a concrete surface and analyze the width of the detected crack.

When cracks are filled by the self-healing repair mortar installed on the concrete surface, even though the cracks are normalized, it can be understood that restoration by the mortar has not been performed due to some grooves formed due to differences in materials.

The monitoring device of the present invention can take measures to distinguish a part restored by mortar from a crack by performing the process of setting, binarization, extraction, determination, and analysis.

According to the monitoring device of the present invention, the monitoring process of crack restoration of the self-healing repair mortar, which was performed with the naked eye, can be automated.

1 is a schematic diagram showing a monitoring device of the present invention.
2 is a schematic diagram showing an assay unit.
Figure 3 is a schematic diagram showing the operation of the analysis unit.
4 is a schematic diagram showing the operation of the analyzer.
5 is a flowchart showing the monitoring method of the present invention.
6 is a schematic diagram showing a crack reference table of the first experimental example.
7 is a chart for calculating the crack width of the first experimental example.
8 is a schematic diagram showing a second experimental example.
9 is a schematic diagram showing the state after 75 days of age in the second experimental example.
10 is a chart showing a comparison result between the crack width analyzed by the monitoring device and the measured crack width.
11 is a chart comparing crack width analysis results after 75 days of age with initial cracks.

Hereinafter, with reference to the accompanying drawings, embodiments of the present invention will be described in detail so that those skilled in the art can easily carry out the present invention. However, the present invention may be embodied in many different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

In this specification, redundant descriptions of the same components are omitted.

In addition, in this specification, when a component is referred to as being 'connected' or 'connected' to another component, it may be directly connected or connected to the other component, but another component in the middle It should be understood that may exist. On the other hand, in this specification, when a component is referred to as 'directly connected' or 'directly connected' to another component, it should be understood that no other component exists in the middle.

In addition, terms used in this specification are only used to describe specific embodiments and are not intended to limit the present invention.

Also, in this specification, a singular expression may include a plurality of expressions unless the context clearly indicates otherwise.

In addition, in this specification, terms such as 'include' or 'having' are only intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, and one or more It should be understood that the presence or addition of other features, numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Also in this specification, the term 'and/or' includes a combination of a plurality of listed items or any item among a plurality of listed items. In this specification, 'A or B' may include 'A', 'B', or 'both A and B'.

Also, in this specification, detailed descriptions of well-known functions and configurations that may obscure the subject matter of the present invention will be omitted.

1 is a schematic diagram showing a monitoring device 100 of the present invention.

The monitoring device 100 of the present invention may include a recording unit 110, a preprocessing unit 130, and an analysis unit 150.

The photographing unit 110 may acquire a photographed image of the concrete surface.

The photographing unit 110 may include photographing means such as a camera and machine vision for photographing the concrete surface. The photographing unit may generate a photographed image of the concrete surface and output it to the preprocessing unit 130 or the analysis unit 150 .

Machine vision can digitally control focus, zoom lens, pan tilt, and the like. Machine vision can photograph concrete cracks through precise control, and can also provide photographed images to the remote analysis unit 150 using Wi-Fi communication.

Alternatively, the photographing unit 110 may include a communication module for wired/wireless communication with a separate photographing unit. The photographing unit 110 may receive a photographed image from the photographing unit through a communication module and transmit the photographed image to the preprocessing unit 130 or the analysis unit 150 .

The pre-processing unit 130 may pre-process the photographed image.

The processing unit may perform pre-processing such as brightness, blur, superimposition, and noise removal using various filtering methods. For example, a median filter may be provided in the preprocessing unit 130 to remove at least one of a shadow and noise included in a photographed image. The median value filter ensures uniform brightness of the captured image and can highlight only cracked areas. Therefore, the median value filter can easily guide the process of extracting crack areas for a captured image in which shadows and noise exist.

The analysis unit 150 may detect and analyze cracks through analysis of photographed images. For example, the analysis unit 150 may analyze a captured image that has passed through a median filter.

2 is a schematic diagram showing the assay unit 150. 3 is a schematic diagram showing the operation of the analysis unit 150.

The analysis unit 150 may include a setting unit 151, a binary unit 153, an extraction unit 155, a determination unit 157, and an analysis unit 159.

The setting unit 151 may set a region of interest (ROI) in the captured image. The region of interest ROI may have a size equal to or smaller than that of the photographed image. The region of interest ROI may be selected by a user or by a machine-learned ROI selection model. A region of interest ROI may include a fissure region. In (a) of FIG. 3, 'Original image' may indicate a region of interest ROI selected by a user.

The binary unit 153 may convert the region of interest into a binarized image d. The binarization unit 153 divides the ROI image into black and white by dividing the ROI image above and below the middle value of gray scale and binarizing it (binarized image d). At this time, the binary part 153 may indicate a crack in white color, like 'Thresholding' shown in (b) of FIG. 3 .

The extraction unit 155 may apply a morphology technique to the binarized image d so that the first candidate region n1 of the crack having continuity and blobs removed is detected.

3 (c ), a first candidate region n1 expected to be a crack, such as 'morphological', may be detected. Here, the morphology technique is a process of manipulating the morphological aspect of an image and can be used as a tool for extracting useful elements for expressing the shape of a region such as a boundary, an attack, or a block.

The determiner 157 may determine the second candidate region n2 by logical ANDing the image including the first candidate region n1 detected by the extractor 155 with the binary image d. The determination unit 157 may correct a phenomenon in which some actual cracks are excluded in the process of applying the morphology technique. Looking at 'Crack detection' in (d) of FIG. 3 corresponding to the image in which the second candidate region n2 is determined, compared to 'morphological' in (c) of FIG. 3 corresponding to the image including the first candidate region n1, the two The shape of the split crack is well represented.

As described above, crack detection may be performed using morphology calculation of the analysis unit 150 after the preprocessing algorithm is applied through the preprocessing unit 130 . However, in this case, since only the crack candidate region (first candidate region n1 or second candidate region n2) can be determined, the crack region and noise are mixed. The analysis unit 159 may be used to track the actual denoised crack.

The analysis unit 159 may track the crack in the second candidate region n2 determined by the determination unit 157 and calculate the width of the tracked crack. Crack width can be an important factor in monitoring repair by self-healing repair mortar.

4 is a schematic diagram showing the operation of the analyzer 159.

The analysis unit 159 may be provided with a tracking means for tracking cracks and a calculation means for calculating crack widths.

The tracking unit may set a detection reference point in the second candidate region n2 of the crack. The tracking unit may search for a region around the detection reference point and perform crack tracking in a direction having a high weight. The tracking unit may extract a vector for the progress path while repeating the process of tracking the crack.

For example, as disclosed in (a) 'Mask & Centerline' of FIG. 4 , the tracking means may mask the crack in a plurality of regions while following the propagation path of the crack. The mask area m is formed in a rectangular shape and may have a size smaller than that of the ROI image. The tracking means can extract the center point of each mask area m. The central point at this time may correspond to the detection reference point. The tracking unit may form a center line o connected to a plurality of center points extracted for each mask area m.

As disclosed in (b) 'Crack width' of FIG. 4 , the calculation unit may search (edge detection) the boundary line of a crack existing in the mask area m.

The calculation means can determine the distance between the vertical line v perpendicular to the center line o and the contact points p1 and p2 of the boundary lines e1 and e2 of the crack as the width of the crack.

The calculation means uses the number of pixels between the intersection of one boundary line e1 and the vertical line v or the first contact point p1 corresponding to the contact point, the intersection point of the other boundary line e2 and the vertical line v, and the second contact point p2 corresponding to the contact point and the resolution of each pixel Crack width can be calculated. For example, the calculator may calculate the width of the crack by multiplying the number of pixels by the resolution of the pixel.

5 is a flowchart showing the monitoring method of the present invention.

The monitoring method shown in FIG. 5 may be performed by the monitoring device 100 of FIG. 1 or the analysis unit 150 of FIG. 2 .

The monitoring method may include a setting step (S 510), a binarization step (S 520), an extraction step (S 530), a determination step (S 540), and an analysis step (S 550).

In the setting step (S510), a Region of Interest (ROI) may be set in the photographed image of the concrete surface. The setting step (S510) may be performed by the setting unit 151.

In the binarization step (S520), the region of interest may be converted into a binarized image. The binarization step (S520) may be performed by the binary unit 153.

In the extraction step (S530), a morphology technique may be applied to the binarized image so that a first candidate region of a crack having continuity and blobs removed is detected. The extraction step (S530) may be performed by the extraction unit 155.

In the determining step ( S540 ), a second candidate region may be determined by ANDing an image including the first candidate region with a binary image. The decision step (S540) may be performed by the decision unit 157.

In the analysis step ( S550 ), cracks may be tracked in the second candidate region determined in the determination step, and the width of the tracked crack may be calculated. The analysis step (S560) may be performed by the analysis unit 159.

6 is a schematic diagram showing a crack reference table of the first experimental example.

In-house tests were conducted to verify the algorithm of crack detection. In the indoor test, after installing the camera at a distance as much as the observation distance from the wall with the crack reference table, 0.1mm, 0.2mm, and 0.5mm virtual cracks (horizontal segments in FIG. 6) were measured, and the reliability of the analysis results was confirmed. The crack reference table was repeatedly photographed 5 times, and the experiment was conducted at a magnification of 32 times and an observation distance of 2.4 m. At this time, the resolution was 0.151mm/pixel.

In the photographed image, as shown in (b) of FIG. 6, cracks were highlighted by making the brightness uniform through a preprocessing process. In addition, cracks were detected using binarization and morphology techniques. For the calculation of crack width, the maximum crack width was measured for each masking image, and finally, the maximum crack width analyzed in the corresponding crack was selected and analyzed conservatively so as not to underestimate the crack width in the field application in the future.

7 is a chart for calculating the crack width of the first experimental example.

Looking at FIG. 7 where the crack width was calculated, the analysis results for the crack widths of 0.5, 0.2, and 0.1 mm showed an error rate of 3.59 to 11.13% with an average of 0.515, 0.212, and 0.102 mm. Looking at the error rate, it may seem that the crack extraction performance and reliability are low. However, in the case of 0.2 and 0.1mm, which showed error rates of 5.71 and 11.13%, the error that may occur when measuring cracks using a crack mirror at the actual inspection site is around 0.05mm. It can be seen that the difference between the actual crack width and the calculated crack width of 0.013 to 0.019 mm is within the actual measurement error range of 0.05 mm. In addition, when looking at the standard deviation and coefficient of variation, the coefficient of variation tended to increase as the crack width decreased, but since the coefficient of variation was stable within 3.5%, it was judged that a certain level of reliability was secured.

As a result of the indoor test, it was evaluated that the crack monitoring device 100 using the image processing technique can not only automatically detect cracks in actual concrete structures, but also secure accuracy within the tolerance range of 0.05 mm for crack width.

8 is a schematic diagram showing a second experimental example.

In order to verify the field applicability and reliability of the automatic concrete crack monitoring device 100, a field verification test was performed on a protective wall constructed with self-healing repair mortar.

In the field test, self-healing repair mortar was installed on the concrete protective wall of the Jeongneungcheongyo ramp section located on the inner ring road in Seoul, and cracks that occurred after repair were detected using a crack monitoring system, as well as cracks over time after cracks occurred. After healing, it was confirmed whether the crack width was reduced.

After self-healing repair mortar construction, naturally occurring cracks in the protective wall were investigated through visual inspection, crack width was measured using a crack microscope, and crack detection and crack width were automatically calculated from the crack image through the monitoring device 100.

After 7 days of repair work, the initial image of cracks generated through on-site inspection was taken, and after 75 days of age, the cracks were re-photographed to confirm the change in crack width.

The result of performing crack analysis on the crack image taken at the first inspection (7th day of age) is shown in FIG. 8, and the result of detecting the crack width is shown in the diagram of FIG. As shown in (c) of FIG. 8, in addition to cracks, fine surface defects and foreign substances present on the surface such as dust, bubbles, foreign substances, and shadows are also analyzed as cracks in the concrete photographed image. Therefore, after performing binarization and morphology techniques to remove micro-defects, it can be confirmed that only cracks were detected as shown in FIG.

Cracks in concrete show the characteristic of having different crack widths at each point even within the same continuous crack. In fact, as a result of measuring the crack width using a cracking microscope, it was observed that the crack width was distributed in various ways, ranging from 0.1 to 0.3 mm. In addition, even in the case of a crack microscope used to measure the crack width, measurement errors may occur, and since it is practically impossible to measure cracks at exactly the same point as the monitoring device 100, a true value such as a virtual crack in an indoor test cannot be defined. Therefore, in this field test, it is considered that it is possible to verify a system that roughly compares the error rate of the crack width analyzed through the developed monitoring device 100 technology and compares the initial crack and the crack width after healing of the self-healing repair mortar.

As shown in FIG. 8(f), the crack width was analyzed for three points in the crack detected, and the result of comparison with the measured crack width is shown in FIG. 10.

The error rate of 1.20~15.18% compared to the actual crack width was found to be rather high, but this is thought to be due to the inability to accurately match the error and comparison position during crack microscopy measurement as described above, and the measurement range of crackers generally used for field inspection Since the difference is within 0.05 mm of phosphorus, it was evaluated that it was possible to detect actual concrete cracks using the developed system and secure the crack width with a certain level of accuracy.

9 is a schematic diagram showing the state after 75 days of age in the second experimental example.

The initial crack was photographed, and after 75 days of age, the image was retaken and the crack width was measured for the same crack. The image preprocessing and crack detection process were performed in the same way as the initial crack measurement method, and the crack detection result is shown in FIG. 9 .

Compared to the initial crack, the reduction in crack width can be clearly confirmed, and it can be seen that the cracks at some points are completely healed. Looking at the diagram in FIG. It was confirmed that % healed, and points 2 and 3 were analyzed to have healed 60.04% and 83.66% cracks, respectively, compared to the initial cracks. do.

Meanwhile, the embodiments of the present invention are not implemented only through the devices and/or methods described so far, and may be implemented through a program that realizes functions corresponding to the configuration of the embodiments of the present invention or a recording medium in which the program is recorded. And, such implementation can be easily implemented by those skilled in the art from the description of the above-described embodiment.

Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements of those skilled in the art using the basic concept of the present invention defined in the following claims are also provided. fall within the scope of the invention.

100 ... monitoring device 110 ... shooting unit
130 ... pre-treatment unit 150 ... analysis unit
151 ... setting part 153 ... binary part
155 ... extraction part 157 ... crystallization part
159 ... analysis department

Claims (7)

A photographing unit that acquires a photographed image of the concrete surface;
an analysis unit that detects and analyzes cracks through analysis of the captured image;
A monitoring device comprising a.
According to claim 1,
Further comprising a pre-processing unit for pre-processing the captured image,
The pre-processing unit is provided with a median filter that removes at least one of shadows and noise included in the photographed image,
The analysis unit analyzes the captured image that has passed through the median filter.
According to claim 1,
The analysis unit is provided with a setting unit, a binary unit, an extraction unit, a determination unit, and an analysis unit;
The setting unit sets a region of interest (ROI) in the captured image,
The binary unit converts the region of interest into a binary image,
The extraction unit applies a morphology technique to the binarized image so that a first candidate region of a crack having continuity and blobs removed is detected,
The determiner determines a second candidate region by ANDing an image including the first candidate region with the binarized image;
The monitoring device of claim 1 , wherein the analysis unit tracks cracks in the second candidate region determined by the determination unit and calculates a width of the tracked crack.
According to claim 1,
The analysis unit is provided with a tracking means for tracking the crack,
The tracking means sets a detection reference point in the second candidate region of the crack,
The tracking means searches a region around the detection reference point and proceeds with crack tracking in a direction with a high weight,
The tracking means is a monitoring device for extracting a vector for the progress path while repeating the progress process of the crack tracking.
According to claim 1,
The analysis unit is provided with a tracking means for tracking the crack and a calculation means for calculating the width of the crack;
The tracking means follows the propagation path of the crack and masks the crack as a plurality of regions;
The tracking means extracts the center point of each mask area;
The tracking unit forms a center line connected with a plurality of center points extracted for each mask area,
wherein the calculating means determines a distance between a vertical line perpendicular to the center line and a contact point of a boundary line of the crack as the width of the crack.
According to claim 5,
Wherein the calculating means calculates the width of the crack using the number of pixels between one boundary line and the first contact point of the vertical line and the other boundary line and the second contact point of the vertical line and the resolution of each pixel.
In the monitoring method performed by the monitoring device,
A setting step of setting a region of interest (ROI) in the photographed image of the concrete surface;
a binarization step of converting the ROI into a binarized image;
an extraction step of applying a morphology technique to the binarized image so that a first candidate region of a crack having continuity and blobs removed is detected;
determining a second candidate region by ANDing an image including the first candidate region with the binarized image;
an analysis step of tracking a crack in the second candidate region determined in the determining step and calculating a width of the tracked crack;
A monitoring method comprising a.

Images