KR102026427B1 - Measurement System for concrete crack using smartphone - Google Patents

Abstract

A method for measuring a concrete crack using a smartphone according to the present invention relates to the method for measuring the concrete crack using the smartphone including a distance sensor and a camera to measure a distance, comprising: a step of measuring a distance between the smartphone and a concrete to measure a crack and photographing a crack image with the camera (S100); a step of performing parallel correction as the camera and the concrete surface are not exactly parallel (S200); a step of applying an image processing technique to the photographed image (S300); and a step of determining and evaluating a crack width and a crack length of the concrete (S400).

Description

Measuring method of concrete crack using smartphone {Measurement System for concrete crack using smartphone}

The present invention relates to a method for measuring a concrete crack using a smart phone to measure a photographing distance and concrete crack image using a distance sensor and a smartphone, and to apply an image processing technique to the photographed image.

Cracks on the surface of the concrete structure is the most obvious type of damage that can intuitively determine the damage of the structure, these cracks can increase the user's anxiety and cause corrosion of the reinforcing bars. Therefore, investigating the width of cracks and their corresponding crack lengths is of great importance in terms of structural maintenance and damage assessment. In the safety inspection and precision safety diagnosis of concrete structures, it is essential to investigate the width and length of cracks. Current crack measurement is mainly used by the naked eye evaluation, and equipment such as a crack gauge for measuring the crack width can be used, but is rarely used due to the time constraints of the site. Although accurate measurements of the width and length of cracks are the most basic data for efficient maintenance of structures, visual measurements are considered to be inconsistent results for operators, resulting in significant errors. It is very efficient in terms of time and economics to be able to evaluate crack width and crack length in real time in the field just by taking a crack image using a smartphone.

 The method of identifying the damage of the structure by using the image processing technique is a very important factor in the development of basic technology for disaster safety management and advanced maintenance of concrete structure. Recently, attention to damage status of many structures due to frequent earthquakes and countermeasures for efficient maintenance of structures have emerged as important keywords for future structure management. Since Korea is no longer a safe area for earthquakes, and the need for structure maintenance is increasing, reliable damage status data is required. Among them, the crack measurement is the most basic damage condition analysis stage of the structure, so it is the basis of the damage status and maintenance of the structure.

 Recently, studies to analyze the damage state of structures using drones and image processing techniques have been conducted in foreign countries, but due to the nature of cracks, the width of the structure is not showing reliable research results. Although the field of image processing such as IT technology and virtual reality has been developed in Korea, the application of crack measurement field is still insignificant.

 Measurement of crack width and crack length is essential for the inspection and safety diagnosis of structures. The evaluation of cracks by the naked eye of the operator is less accurate for the results, difficult to measure on-site during the underground structure and night work, and it is difficult to expand the measuring range or speed up the work due to excessive cost and manpower. Therefore, it is necessary to develop measurement equipment with excellent measurement accuracy and operator convenience for various concrete structures.

KR 10-1806272 B1 KR 10-1501120 B1 KR 10-1316451 B1

The present invention to solve the above problems, to provide a concrete crack measuring method using a smart phone excellent in measurement accuracy and measurement convenience of the operator.

Concrete crack measurement method using a smart phone according to the present invention in the concrete crack measurement method using a smart phone including a distance sensor and a camera to measure the distance, the distance between the concrete and the concrete to measure the crack Measuring and taking a crack image with the camera (S100); Performing parallel correction because the camera and the concrete surface are not exactly parallel (S200); And applying the image processing technique to the photographed image (S300) and determining and evaluating the crack width and the crack length of the concrete (S400).

The present invention has the effect of improving the efficiency of maintenance of the structure and reducing the economic cost through the advanced concrete crack management technology, and providing the low-cost, high-efficiency working environment when measuring the damage of the structure by the measurers of the concrete structure Doing.

1 is a flow chart of a concrete crack measurement method using a smart phone according to the present invention.
2 shows concrete that is the target of crack width and crack length evaluation as an example of a concrete crack measurement method using a smart phone according to the present invention.
Figure 3 is a front view of the smartphone used in the concrete crack measurement method using a smartphone according to the present invention.
Figure 4 is a rear view of the smartphone used in the concrete crack measurement method using a smartphone according to the present invention.
5 is an embodiment of a smartphone application screen used in the concrete crack measurement method using a smartphone according to the present invention.
6 is an algorithm of a concrete structure crack width estimation method using an image processing method that can be applied to the concrete crack measurement method using a smart phone according to the present invention.

Hereinafter, although described with reference to the drawings according to an embodiment of the present invention, this is for easier understanding of the present invention, the scope of the present invention is not limited thereto.

When any part of the specification is to "include" any component, this means that it may further include other components, except to exclude other components unless otherwise stated. In addition, the terms "... unit", "module", etc. described in the specification mean a unit that processes at least one function or operation, which may be implemented in hardware or software or a combination of hardware and software. .

Throughout the specification, when a part is "connected" to another part, this includes not only "directly connected" but also "electrically connected" with another element in between. .

Prior to the description, a number of aspects and embodiments are described herein, which are merely illustrative and not limiting.

After reading this specification, skilled artisans will appreciate that other aspects and embodiments are possible without departing from the scope of the invention.

1 is a flow chart of a concrete crack measurement method using a smartphone according to the present invention, Figure 2 shows an example of the concrete of the crack width and crack length evaluation of the concrete crack measurement method using a smartphone according to the present invention, 3 is a front view of the smartphone used in the concrete crack measurement method using a smartphone according to the present invention, Figure 4 is a rear view of the smartphone used in the concrete crack measurement method using a smartphone according to the present invention, 5 is an embodiment of a smartphone application screen used in the concrete crack measurement method using a smartphone according to the present invention, Figure 6 is an image processing technique that can be applied to the concrete crack measurement method using a smartphone according to the present invention Algorithm of crack width estimation method for concrete structures using

Next will be described in detail with respect to the concrete crack measurement method using a smart phone according to the present invention with reference to FIGS.

Concrete crack measurement method using a smart phone according to the present invention in the concrete crack measurement method using a smart phone including a distance sensor and a camera to measure the distance, the distance between the concrete and the concrete to measure the crack Measuring and taking a crack image with the camera (S100); Performing parallel correction because the camera and the concrete surface are not exactly parallel (S200); And applying the image processing technique to the photographed image (S300) and determining and evaluating the crack width and the crack length of the concrete (S400).

In addition, the image processing method is characterized by removing the noise on the background of the photographed image to classify the crack, and remove the noise by determining the threshold value for the binarization as the local current value in consideration of the concrete surface roughness state. .

The step S400 is characterized in that the width of the crack is determined using the histogram for the crack image, and the crack length is determined by labeling the crack image.

An object of the present invention is to measure the shooting distance and concrete crack image using a distance sensor and a smart phone, and to evaluate the crack width and crack length in real time by applying an image processing technique (digital image processing) to the captured image To provide a way.

To describe the crack width and crack length evaluation algorithm in more detail, the measurement method using the image processing technique can be used in various fields such as the analysis of the dynamic characteristics of the structure in an economical and effective way compared to the measurement method using a conventional sensor and equipment. have. As shown in FIG. 2, the image of the crack generated on the surface of the concrete structure is not easily distinguished between the crack and the non-crack by the roughness and dust of the concrete surface. In order to solve this problem, in the conventional image processing technique, a method of classifying cracks by removing noise on a background has been studied.

In the present invention, by considering the surface roughness of the concrete to determine the threshold value for the binarization as a regional statistical value to remove the basic noise, and to determine the width of the crack using the histogram for the crack image. In addition, the crack length is determined by labeling the crack image. The OpenCV library provides many functions of this image processing technique, which can be implemented on Java and Android platforms.

Next, the shooting distance measuring module will be described in more detail. Using a smartphone, an image of a crack generated on the surface of a concrete structure is taken, and in order to evaluate the width and length of the crack, the distance from the smartphone camera to the crack is evaluated. You need to know the distance, and correction is required because the smartphone's camera and the concrete surface are not exactly parallel. Therefore, there is a need for a module that can measure the shooting distance of a smartphone using a distance sensor and Arduino. The distance measurement module controls four distance sensors on the motherboard, and the measured distances are wirelessly transmitted to the application of the smartphone.

In addition, for smartphone applications, algorithms developed in the Java language are ported to the Android language and applied to smartphone applications. The measurement distance received from the shooting distance measuring module is used for the evaluation of the crack width and the crack length in the application of the smartphone.

The present invention provides a method for measuring the surface crack width and crack length of a concrete structure in real time using a laser or an ultrasonic sensor and a smart phone. To this end, the following method is recommended when implementing the present invention. .

 An image processing technique using OpenCV, a module consisting of a distance sensor and Arduino, and an algorithm for evaluating the crack width and crack length by applying the image processing technique to the crack image are applied. Android, the coding language of smartphones, is based on the Java language, so the development of algorithms is recommended to be coded in the Java language. Image processing for the crack image can be performed by registering the OpenCV library to the Java platform. Gray image conversion and binarization processes are basically applied. The threshold selection, which is an essential step in the binarization process, has already been performed in the conventional study based on the relevant theories. In the present invention, the threshold is determined through a regional statistical approach based on the error analysis. do.

In addition, conventional studies related to the evaluation of concrete crack widths calculate the number of pixels directly after detecting only cracks from the background. This method is not suitable given the situation where the crack width must be evaluated quickly in the field. Therefore, in the present invention, the histogram analysis is first performed in connection with the measurement of the crack width, and the crack width is evaluated by calculating the crack width at the maximum pixel position with respect to the axis. At this time, the scale is considered between the number of pixels and the actual crack width of the image based on the photographing distance measured from the laser or the ultrasonic sensor. In connection with the crack length measurement, labeling of the crack image is performed to evaluate the crack length for each crack object. In this step, the selection of the threshold by the statistical approach should be reliable for the evaluation of the crack width by the image processing technique, and it is a reliable error when comparing the crack width by the image processing technique with the crack width measured directly by the crack gauge. Try to have a value in the range.

Next, we apply a system (application) that can measure the distance from the smartphone to the concrete surface using a laser or ultrasonic sensor and Arduino. The type of sensor is selected by experimenting with laser or ultrasonic sensor in consideration of the roughness of the concrete surface and the site situation. In addition, as shown in Figures 3 to 4, the smartphone and the sensor and the power supply device is configured as a system to be easily carried by applying a smartphone mounting module and a frame using a 3D printer to ensure the accuracy of the sensor, The stability of the shooting distance measuring system should be checked and applied in consideration of the voltage.

Next, as shown in FIG. 5, a smartphone application is produced on the Android platform, and an algorithm developed in the Java language is applied to the application. In addition, a wireless communication function is implemented by incorporating a photographing distance measuring device, and errors can be minimized through debugging. Finally, smartphones, sensors, motherboards, power supplies, etc. are attached to the fabricated frame to form the final system and connect the test process. In this step, the frame produced by the 3D printer is simplified through various modifications, and the field applicability is examined through field experiments.

Next will be described in detail with respect to the concrete structure crack width evaluation method that can be applied to the concrete crack measurement method using a smart phone according to the present invention with reference to FIG.

In the method for evaluating the crack width of the concrete structure, Step of photographing the concrete surface with a camera (S10); Converting the captured image into a gray image (S20); Performing smoothing through a bilateral blur to remove noise generated in the captured image by the concrete surface (S30); Applying an Adaptive Threshold Gaussian Constant to remove the effects of shadows and light reflections on the captured image (S40); Applying a Closing operation (Dilation-Erosion) to compensate for the effect of separating the continuous concrete crack object due to the Adaptive Threshold Gaussian Constant (S50) and to detect the uniform candidate region in the binarization and Closing operation applied image And applying a labeling algorithm (S60).

In addition, the step (S30) may apply a diameter of 15 for the bilateral blur, and a standard deviation of 40 for color and space.

In addition, the step S40 may be applied to Block Size 7, Constant C 8.0.

The following describes the Bialateral Blur in detail.

Bilateral Blur is one of the image analysis methods known as edge-preserving smoothing and performs smoothing using a bidirectional filter. Bilateral Blur obtains a weighted average from each pixel and surrounding elements, and the weight has two components. One is the same as the weight used in Gaussian Smoothing, and the other is similar to the Gaussian weight, but uses the weight determined by the brightness difference from the center pixel value, not the value determined by the distance from the center. In other words, Bilateral Blur can be regarded as Gaussian Smoothing, which gives greater weight to similar pixels, which can be effective for removing point-shaped noise on concrete surfaces. Bilateral Blur is defined as in Equation 1 below.

(Equation 1)

Here, f and h mean that the input image and the resultant image are multi-band, k is defined as Equation 2 as Kernel.

(Equation 2)

The following describes the Adaptive Threshold in detail.

Adaptive Threshold is a method of performing binarization by analyzing the distribution of surrounding pixels for a specific region and determining a threshold value. Adaptive Threshold T (x, y) has a different value for a specific region. T (x, y) is determined by subtracting a specified constant from a weighted average value calculated at each block size × block size. The Adaptive Threshold method is useful when the pixel values change gradually because the image contains strong lighting or reflections.In the case of the Adaptive Threshold Mean Constant, the weights are all equal, and in the case of the Adaptive Threshold Gaussian Constant, the weights are Gaussian functions. It is specified in the form of to give more weight to the center pixel. In the present invention, it is recommended to apply the Adaptive Threshold Gaussian Constant to remove the stain and shadow effects on the concrete image. Adaptive Threshold is calculated according to Equation 3 below.

(Equation 3)

The following describes the dilation operation in detail.

The dilation operation is the convolution of a portion of the image with the kernel. The kernel is similar to a template or mask, and the dilation operation has the effect of selecting a local maximum. The maximum pixel value for a specific region is obtained from part of the image and the kernel's computational domain, and the pixel value below the kernel's fixed point is set to this maximum value. As a result, the object of interest in the image is expanded. The dilation operation may be applied to labeling through the connected component, and is calculated through Equation 4 below.

(Equation 4)

The following describes the Erosion operation in detail.

Erosion operation is the opposite of Dilaton. Erosion calculates the local minimum under the kernel. Erosion operations can be used to remove speckle noise from images. Speckles are removed through the Erosion operation, and the regions of interest remain unaffected. Erosion operation is calculated according to the following equation (5).

(Equation 5)

Labeling for binarized images has been applied in many fields for image processing. When the input image includes one or more connected elements, the background pixel value 0 may be detected from the matrix X ₀ having the same size as the input image A. Except for the position corresponding to X ₀ thus detected, each connected element of the input image A becomes a foreground pixel value. The labeling operation starts with X ₀ for the connected elements and can find all connected elements by operating over the entire range of the input image A. The labeling operation is calculated by the following equation (6).

(Equation 6)

Here, B is any structural element to the associated component is detected, the algorithm of the equation (6) is terminated when about X _k containing all the connected elements of an input image A X _k = X _k-1 is satisfied.

Next, a method of detecting a crack candidate region using a labeling area ratio will be described in detail.

Concrete cracks have a characteristic of maintaining the direction of progress with respect to the shape of the cause of occurrence. Considering these characteristics, the area ratio of the labeled area can be divided into crack and non-cracked area. The area ratio with respect to the labeled area may be calculated by Equation 7 as follows.

(Equation 7)

Here, A _ob is the object area of each area detected by labeling, and A _cb is the rectangular area of each area. The area ratio F _cb can have a value between 0 and 1.0, and in general, the value becomes smaller in the case of independent cracks.

The following describes the crack width estimation using the histogram in detail.

Histogram is a simple summation of data accumulated on a fixed number of axes. Histogram is the sum of the number of features that appear from the size, direction, and color of a gradient, and shows a statistical plot of a given data distribution. Histograms generally consist of orders of magnitude lower than the original data. The histogram may indicate crack pixel frequencies on the x and y axes in the labeled crack candidate region as shown in FIGS. 1 and 2. In FIG. 1 and FIG. 2, the maximum Histogram frequency and its position for each axis may be used for crack width evaluation using FIG. 3 and Equation 8 below.

,

,

,

,

(Equation 8)

Where w is the maximum crack width for each crack candidate region and w ₁ and w ₂ are the crack widths for the histogram maximum frequency on the x and y axes. Δx ₁ is the maximum frequency based on the x-axis, and Δy ₁ is the y-axis frequency at the Δx ₁ position. Δy ₂ is the maximum frequency based on the y-axis, and Δx ₂ is the x-axis frequency at the Δy ₂ position.

Recently, the number of damage to concrete structures is increasing due to frequent earthquakes in Korea, and the crack damage of concrete structures has become a social problem due to such earthquake damage. Therefore, the development of an algorithm that can evaluate crack width and length using an image processing technique and the development of a crack measurement system are becoming more important.

 The present invention relates to the development of advanced core technology and to secure intellectual property rights for the measurement of crack width and crack length, and by the development of such a crack measuring device technology, the general public can easily measure cracks, thereby relieving some of the anxiety caused by earthquake damage. have. Improved accuracy of crack measurement can improve national disaster prevention planning and management techniques for concrete structures, and can lead to the activation of related industries by developing crack measurement and operation technology by high-tech convergence. In addition, by improving the concrete crack management technology, it is possible to improve the efficiency of structural maintenance and thereby reduce the economic cost. These technologies can be utilized as core equipment in national disaster facility management projects such as the Ministry of Land, Infrastructure and Transport and the Ministry of Public Administration and Security by providing real-time crack measurement data linked to the standardization of crack width measurement and integrated DB. They will provide a low cost, high efficiency working environment when assessing damage to structures.

Although described above with reference to the drawings according to an embodiment of the present invention, those of ordinary skill in the art will be able to perform various applications and modifications within the scope of the present invention based on the above contents.

Claims (3)

In the concrete crack measurement method using a smart phone comprising a distance sensor and a camera to measure the distance,
Measuring a distance between the smartphone and concrete to measure cracks and taking a crack image with the camera (S100);
Performing parallel correction because the camera and the concrete surface are not exactly parallel (S200);
Applying an image processing technique to the captured image (S300) and
Determining and evaluating the crack width and crack length of the concrete (S400),
The image processing technique is to remove the noise on the background of the photographed image to classify the crack, and to determine the threshold value for the binarization as a local communication value in consideration of the concrete surface roughness state, to remove the noise,
The step (S400),
Use the histogram for the crack image to determine the width of the crack, labeling the crack image to determine the crack length,
Method for evaluating the crack width of the concrete in the step (S400),
Photographing the concrete surface with a camera (S10); Converting the captured image into a gray image (S20); Performing smoothing through a bilateral blur to remove noise generated in the captured image by the concrete surface (S30); Applying an Adaptive Threshold Gaussian Constant to remove the effects of shadows and light reflections on the captured image (S40); Applying a Closing operation (Dilation-Erosion) to compensate for the effect of separating the continuous concrete crack object due to the Adaptive Threshold Gaussian Constant (S50) and to detect the uniform candidate region in the binarization and Closing operation applied image Concrete crack measurement method using a smartphone comprising a step of applying a labeling algorithm (S60).
delete delete

Images