A method for detecting the spacing of square reinforcement cage structures of an automated diaphragm wall

Abstract

The application discloses a kind of detection methods of automatic processing diaphragm wall square reinforcement cage structure spacing, first acquisition reinforcement cage image, only two stirrups and two straight bars are contained in each reinforcement cage image, and straight bar is at certain angle with image length and width direction;Then, reinforcement cage image is preprocessed;Finally, reinforcement cage image is segmented into multiple sub-block images of same size, extract straight line in sub-block image, obtain the slope and intercept of each straight line;The slope of all straight lines is clustered, and the cluster centers k1 and k2 are obtained;Select multiple straight lines from all straight lines as selected straight line, the longitudinal intercept of all selected straight lines is clustered, and the cluster centers b3 and b4 are obtained, with the cluster center k1 as the slope, the cluster center b3 and b4 as the intercept, obtain two straight line equations, and then obtain the reinforcement spacing D1 of reinforcement cage;Similarly, the reinforcement spacing D2 of reinforcement cage is obtained, and the straight bar spacing and stirrup spacing are determined.The method realizes the automatic detection of reinforcement spacing in reinforcement cage, and has high processing efficiency and detection accuracy.

Description

Technical Field

[0001] This invention belongs to the field of quality inspection technology for steel cages in automated processing, specifically a method for detecting the spacing of square steel cage structures for diaphragm walls in automated processing. Background Technology

[0002] With the continuous development of intelligent construction technology, diaphragm wall reinforcement cages have transitioned from manual binding to automated processing. Automated equipment enables automatic welding of straight bars and stirrups, significantly improving processing efficiency. Diaphragm wall concrete structures play a major load-bearing role in buildings, playing a crucial role in their stability and safety. Reinforcement cages are a method of steel reinforcement configuration in diaphragm wall concrete structures, primarily serving a tensile function to increase the tensile strength of the concrete. The spacing of the straight bars and stirrups in the reinforcement cage affects the quality of the finished product, thus impacting the load-bearing capacity of the diaphragm wall concrete structure. Inadequate spacing of a single straight bar or stirrup can lead to defects in the entire finished reinforcement cage. Therefore, it is necessary to inspect the spacing of the straight bars and stirrups during processing.

[0003] Currently, the detection of straight bar spacing and stirrup spacing during automated rebar processing is still in its initial stage, lacking automated detection methods. Therefore, this invention proposes a method for detecting the spacing of square rebar cages used in automated diaphragm wall construction, which is of great significance for real-time quality inspection of rebar cage processing. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the technical problem this invention aims to solve is to provide an automated method for detecting the spacing of square steel cage structures used in diaphragm wall construction.

[0005] The technical solution adopted by the present invention to solve the aforementioned technical problem is as follows:

[0006] A method for detecting the spacing of a diaphragm wall square steel cage structure using automated processing, characterized by the following steps:

[0007] Step 1: Acquire images of the rebar cage. Each image of the rebar cage contains only two stirrups and two straight bars, and the straight bars form a certain angle with the length and width directions of the image.

[0008] Step 2: Preprocess the steel cage image;

[0009] Step 3: Divide the rebar cage image into multiple sub-block images of the same size, and each rebar cage image will result in a sub-block image array;

[0010] Extract all straight lines from the sub-block image to obtain the slope k, x-intercept and y-intercept b1 of each line; cluster the slopes of all lines to obtain the cluster centers k1 and k2.

[0011] Let Δk = |k - k1|, select lines where Δk is less than the slope difference threshold, and choose multiple lines with smaller Δk from these lines as the selected lines; map the ordinate b1 of all selected lines from the sub-block image coordinate system to the rebar cage image coordinate system, and cluster the ordinates of all selected lines in the rebar cage image coordinate system to obtain cluster centers b3 and b4, and then obtain the equations of the two lines corresponding to the cluster center k1 as follows:

[0012]

[0013] The spacing D1 of the reinforcing bars in the reinforcing cage is:

[0014]

[0015] Let Δk = |k - k2|, and repeat the above process to obtain the equations of the two lines corresponding to the cluster center k2:

[0016]

[0017] The spacing D2 of the reinforcing bars in the reinforcing cage is:

[0018]

[0019] In the formula, b5 and b6 are the cluster centers obtained by the y-intercepts of all selected lines;

[0020] Let q be the design value of stirrup spacing, and M be the stirrup spacing judgment threshold. If |D1-q|≤M, then D1 is determined to be stirrup spacing and D2 is straight bar spacing; otherwise, D1 is straight bar spacing and D2 is stirrup spacing.

[0021] Alternatively, let q be the design value of the straight bar spacing and M be the threshold for judging the straight bar spacing. If |D1-q|≤M, then D1 is determined to be the straight bar spacing and D2 is the stirrup spacing; otherwise, D1 is the stirrup spacing and D2 is the straight bar spacing.

[0022] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0023] 1. Divide each steel cage image into multiple sub-block images and extract straight lines from the sub-block images. The sub-block images themselves are small, which can greatly improve the image processing efficiency. Moreover, the steel bars occupy a large area in the sub-block images, which is convenient for straight line extraction and helps to improve the detection accuracy of straight bars and stirrup spacing.

[0024] 2. This invention achieves automated detection of the spacing between straight bars and stirrups in a reinforcing cage. It employs an industrial camera for image acquisition, with the shooting frequency matching the movement frequency of the reinforcing cage on the production line. This ensures that the image contains only one stirrup spacing, guaranteeing detection accuracy. This method allows for comprehensive quality inspection of the entire reinforcing cage without blind spots and can deduce the specific location of unqualified spacing based on the movement and shooting frequencies, thus locating defects.

[0025] 3. This invention significantly reduces image memory usage through image grayscale processing, effectively improving the speed of computer image processing and enabling real-time quality inspection. It allows for rapid identification of defects. During image acquisition, a white background board is placed under the steel cage to maximize the distinction between the background and foreground, facilitating foreground and background segmentation. Attached Figure Description

[0026] Figure 1 This is an overall flowchart of the present invention;

[0027] Figure 2 This is an example image of the steel cage of the present invention. Detailed Implementation

[0028] Specific embodiments are given below with reference to the accompanying drawings. These specific embodiments are only used to further illustrate the technical solutions of the present invention, but are not intended to limit the scope of protection of this application.

[0029] This invention provides an automated method for detecting the spacing of square steel cage structures with diaphragm walls (hereinafter referred to as the method, see [link]). Figures 1-2 This includes the following steps:

[0030] Step 1: Use an industrial camera to acquire images of the rebar cage. The acquisition frequency of the industrial camera is the same as the movement frequency of the rebar cage on the production line to ensure that each image of the rebar cage contains only two stirrups and two straight bars. In the image of the rebar cage, the straight bars are at a certain angle to the x and y axes of the rebar cage image coordinate system. The origin of the rebar cage image coordinate system is located at the lower left corner of the rebar cage image, and the x and y axes are along the length and width of the rebar cage image, respectively.

[0031] An industrial camera is mounted above the rebar cage production line via a gantry frame. It is necessary to ensure that the two straight bars and the two adjacent stirrups of the rebar cage are all within the shooting range of the industrial camera. The camera is perpendicular to the shooting area, that is, perpendicular to the plane where the upper edges of the two stirrups are located. A white background is placed in the shooting area to increase the difference between the foreground and background of the image, which facilitates image processing. In this embodiment, the angle between the straight bars in the rebar cage image and the x and y axes of the rebar cage image coordinate system is ±45 degrees.

[0032] Step 2: Preprocess the steel cage image, including denoising, grayscale conversion, and enhancement.

[0033] Since the steel cage itself is dark in color, using a color model has a better noise reduction effect. The images captured by the industrial camera are all in RGB format. Compared with the RGB color model, the Lab color model has better perceptual balance. It is mainly composed of one luminance component (L) and two color components (a, b). The value range of the luminance component L is [0, 100]. The larger the value, the higher the luminance. When L is 100, it is white, and when L is 0, it is black. The color component a represents the change from green to red, and b represents the change from blue to yellow. The value range of color components a and b is [-128, 127].

[0034] The luminance component L and color components a and b of each pixel in the steel cage image are calculated using equation (1).

[0035]

[0036] In equation (1), R, G, and B are the R, G, and B values ​​of a pixel in RGB format, and (X, Y, Z) are the coordinates of the pixel in the intermediate coordinate system; X n Y n Z n All are coefficients, with default values ​​of 95.047, 100.0, and 108.883 respectively; f(t) is an intermediate function, and t takes values ​​of... and

[0037] The Lab color value H of the pixel is obtained by weighted summation of the luminance component L and color components a and b of the pixel using equation (2).

[0038] H=α×L+β×a+γ×b (2)

[0039] In the formula, α, β, and γ are all weights, and α + β + γ = 1;

[0040] The Lab color value H of a pixel is compared with the segmentation threshold H0. If the Lab color value H of a pixel is greater than or equal to the segmentation threshold H0, the pixel belongs to the foreground; if the Lab color value H of a pixel is less than the segmentation threshold H0, the pixel belongs to the background, i.e., it belongs to noise. The R, G, and B values ​​of the pixel are all modified to 255. All pixels of the steel cage image are traversed to separate the foreground and background of the steel cage image in order to remove noise.

[0041] Since the denoised steel cage image itself is in RGB format, it occupies a large amount of memory and will affect the processing efficiency. Therefore, the denoised steel cage image is grayscaled by equation (3) to obtain the grayscale steel cage image.

[0042] I=0.2126R+0.7152G+0.0722B (3)

[0043] In the formula, I is the gray value of the pixel;

[0044] Histogram equalization was used to enhance the grayscale image of the steel cage, thereby expanding the grayscale levels with a large range of similar grayscale values ​​and reducing the grayscale levels with a small range of similar grayscale values.

[0045] Step 3: Calculate the spacing of the reinforcing bars in the steel cage, and determine the spacing of straight bars and stirrups based on the threshold values;

[0046] The size of the image itself also affects the processing efficiency. Therefore, the preprocessed steel cage image is divided into multiple sub-block images of the same size. Each sub-block image corresponds to a position in the steel cage image. This position mapping relationship can form a sub-block image array from all the sub-block images obtained from a preprocessed steel cage image, which is divided into n rows and m columns.

[0047] The origin of the sub-block image coordinate system is set at the bottom left corner. The x and y axes of the sub-block image coordinate system are along the length and width directions of the sub-block image, respectively. The Hough transform is used to extract all straight lines in the sub-block image. The equations of the Hough transform lines are as follows:

[0048] ρ=x cosθ+y sinθ (4)

[0049] In the formula, (x,y) represents the coordinates of any point on the line in the sub-block image coordinate system, (ρ,θ) represents the coordinates of any point on the line in the polar coordinate system, and θ is the angle between the line connecting any point on the line and the origin of the sub-block image coordinate system and the x-axis of the sub-block image coordinate system.

[0050] The Hough transform equations of each line are converted into slope-intercept equations to obtain the slope k, horizontal intercept a1, and vertical intercept b1 of each line. The slope k of all lines is clustered into two classes to obtain the cluster centers k1 and k2. All lines are screened according to the cluster centers k1 and slope k. Let Δk = |k-k1|, select the lines whose Δk is less than the slope difference threshold u, and sort all the selected lines according to Δk from small to large. Keep the p lines with smaller Δk. These lines are called the selected lines. The vertical intercept b1 of the p selected lines is transformed from the sub-block image coordinate system to the steel cage image coordinate system by equation (4) to obtain the vertical intercept b2 of the selected lines in the steel cage image coordinate system. Then the vertical intercept of the selected lines is mapped from the sub-block image coordinate system to the steel cage image coordinate system.

[0051]

[0052] In the formula, (i,j) represents the position of the sub-block image in the sub-block image array, and C1 is the width of the sub-block image;

[0053] Cluster the y-intercepts of all selected straight lines in the coordinate system of the steel cage image, and cluster the y-intercepts into two classes to obtain cluster centers b3 and b4; take cluster centers b3 and b4 as y-intercepts and cluster center k1 as slope to obtain the equations of the two straight lines corresponding to cluster center k1.

[0054]

[0055] The spacing D1 of the reinforcing bars in the reinforcing cage is:

[0056]

[0057] Let Δk = |k - k2|, and repeat the above process to obtain the two linear equations corresponding to the cluster center k2;

[0058]

[0059] The spacing D2 of the reinforcing bars in the reinforcing cage is:

[0060]

[0061] In the formula, b5 and b6 are the cluster centers obtained by the y-intercepts of all selected lines;

[0062] Let q be the design value of stirrup spacing, and M be the stirrup spacing judgment threshold. If |D1-q|≤M, then D1 is determined to be stirrup spacing and D2 is straight bar spacing; otherwise, D1 is straight bar spacing and D2 is stirrup spacing.

[0063] Alternatively, let q be the design value of the straight bar spacing and M be the threshold for judging the straight bar spacing. If |D1-q|≤M, then D1 is determined to be the straight bar spacing and D2 is the stirrup spacing; otherwise, D1 is the stirrup spacing and D2 is the straight bar spacing.

[0064] Any aspects not covered in this invention are applicable to existing technologies.

Claims

1. A method for detecting the spacing of a diaphragm wall square steel cage structure using automated processing, characterized in that, The method includes the following steps: Step 1: Acquire images of the rebar cage. Each image of the rebar cage contains only two stirrups and two straight bars, and the straight bars form a certain angle with the length and width directions of the image. Step 2: Preprocess the steel cage image; Step 3: Divide the rebar cage image into multiple sub-block images of the same size, and each rebar cage image will result in a sub-block image array; Extract all straight lines from the sub-block image to obtain the slope k, x-intercept and y-intercept b1 of each line; cluster the slopes of all lines to obtain the cluster centers k1 and k2. Let Δk = |k - k1|, select lines where Δk is less than the slope difference threshold, and choose multiple lines with smaller Δk from these lines as the selected lines; map the ordinate b1 of all selected lines from the sub-block image coordinate system to the rebar cage image coordinate system, and cluster the ordinates of all selected lines in the rebar cage image coordinate system to obtain cluster centers b3 and b4, and then obtain the equations of the two lines corresponding to the cluster center k1 as follows: The spacing D1 of the reinforcing bars in the reinforcing cage is: Let Δk = |k - k2|, and repeat the above process to obtain the equations of the two lines corresponding to the cluster center k2: The spacing D2 of the reinforcing bars in the reinforcing cage is: In the formula, b5 and b6 are the cluster centers obtained by the y-intercepts of all selected lines; Let q be the design value of stirrup spacing, and M be the stirrup spacing judgment threshold. If |D1-q|≤M, then D1 is determined to be stirrup spacing and D2 is straight bar spacing; otherwise, D1 is straight bar spacing and D2 is stirrup spacing. Alternatively, let q be the design value of the straight bar spacing and M be the threshold for judging the straight bar spacing. If |D1-q|≤M, then D1 is determined to be the straight bar spacing and D2 is the stirrup spacing; otherwise, D1 is the stirrup spacing and D2 is the straight bar spacing.

2. The method for detecting the spacing of automated diaphragm wall square steel cage structures according to claim 1, characterized in that, The formula for calculating the ordinate b2 of the selected line in the coordinate system of the steel cage image is: In the formula, (i,j) represents the position of the sub-block image in the sub-block image array, C1 is the width of the sub-block image, and a1 is the x-intercept of the selected line in the sub-block image coordinate system. The origin of the sub-block image coordinate system is located at the lower left corner of the sub-block image, and the x and y axes are along the length and width directions of the sub-block image, respectively. The origin of the steel cage image coordinate system is located at the lower left corner of the steel cage image, and the x and y axes are along the length and width directions of the steel cage image, respectively.

3. The method for detecting the spacing of automated diaphragm wall square steel cage structures according to claim 1 or 2, characterized in that, The preprocessing includes noise reduction, grayscale conversion, and enhancement.

4. The method for detecting the spacing of automated diaphragm wall square steel cage structures according to claim 3, characterized in that, The noise reduction process includes: Calculate the luminance component L and color components a and b of each pixel in the steel cage image. Perform a weighted sum of the luminance component L and color components a and b of each pixel to obtain the Lab color value of the pixel. If the Lab color value of a pixel is greater than or equal to the segmentation threshold, the pixel belongs to the foreground; otherwise, it belongs to the background. Traverse all pixels in the steel cage image and modify the R, G, and B values ​​of each pixel belonging to the background to 255.

Images