Frame fastness detection method based on high-strength alloy template structure optimization

Abstract

The present invention relates to the field of image analysis technology, and in particular to a method for detecting the tightness of a frame based on the optimization of a high-strength alloy template structure. The method obtains possible weld indicators of the corresponding area block according to the degree of reflection and texture complexity of each area block of the reinforcement welding image, and obtains the amount of defect information of each area block in combination with the difference between the degree of fluctuation of the edge lines in each area block and the extension direction of different edge lines; the fastening quality of the frame of the high-strength alloy template is detected according to the difference in the amount of reflection of the edge lines in the area block of the reinforcement welding image, possible weld indicators and defect information. The present invention performs defect detection based on multiple factors such as the degree of reflection of the area block, the degree of texture complexity, and the degree of fluctuation and extension direction of the edge lines in the area block, thereby improving the accuracy of the detection of the tightness of the frame of the high-strength alloy template.

Description

Technical Field

[0001] The invention relates to the technical field of image analysis, and in particular to a frame fastness detection method based on high-strength alloy template structure optimization. Background Art

[0002] During construction, alloy formwork used for temporary support is mainly made of alloy plates after welding. In the actual construction process, in order to improve the bearing capacity of high-strength alloy formwork under different operating requirements and extend the service life of the formwork, the stability requirements of the alloy formwork under the current construction environment are analyzed, and reinforcements such as reinforcing ribs and corner braces are welded on the formwork frame for reinforcement. The reinforcements are used to enhance the stability of the high-strength alloy formwork structure. The reinforcement capacity of the reinforcement to the formwork frame depends on the quality of the weld between the reinforcement and the formwork frame.

[0003] Cracks at the weld position can significantly reduce the fatigue strength of the welded structure or joint under alternating loads, affecting the quality of the weld between the reinforcement and the template frame. The crack information in the weld area can be analyzed to determine the quality of the alloy template frame reinforcement. The existing method determines the suspected crack pixel points based on the gradient amplitude and grayscale value of the pixel points, and then performs defect detection on the weld surface; however, in order to avoid the protruding part of the reinforcement affecting the assembly between the alloy templates during use, the reinforcement is generally welded in the groove of the high-strength alloy template frame. The alloy template frame will produce shadows under light, resulting in a low accuracy rate for defect detection using only grayscale value and gradient amplitude, which makes the high-strength alloy template frame reinforcement detection inaccurate. Summary of the invention

[0004] In order to solve the technical problem that the shadow produced by the alloy template frame under illumination affects the defect detection, resulting in inaccurate detection of the frame reinforcement of the high-strength alloy template, the purpose of the present invention is to provide a frame fastness detection method based on the optimization of the high-strength alloy template structure, and the technical scheme adopted is as follows:

[0005] The present invention proposes a frame fastness detection method based on high-strength alloy template structure optimization, the method comprising:

[0006] Obtain welding images of reinforcements of high-strength alloy templates;

[0007] Divide the reinforcement welding image into different area blocks; obtain the possible weld index of each area block according to the reflection degree and texture complexity of each area block;

[0008] Obtain edge lines in each area block; obtain defect information of each area block according to the fluctuation degree of all edge lines in each area block and the difference between the extension directions of different edge lines, as well as the possible indicators of the weld;

[0009] The frame fastening quality of the high-strength alloy formwork is detected based on the difference in the amount of reflection of the edge lines in all area blocks of the reinforcement welding image, the possible indicators of the weld and the amount of defect information.

[0010] Furthermore, the method for obtaining possible indicators of the weld of each area block includes:

[0011] Perform edge detection on the reinforcement welding image to obtain the gradient value of each pixel in the reinforcement welding image;

[0012] The variance of the gradient values ​​of all pixels in each area block is used as the texture complexity of each area block; the overall reflectivity of the corresponding area block is obtained according to the grayscale values ​​of all pixels in each area block; and the possible weld index of each area block is obtained according to the texture complexity and the overall reflectivity.

[0013] Furthermore, the method for obtaining edge lines in each area block includes:

[0014] Curve fitting is performed on the edge pixels obtained by edge detection in each area block, and the fitted curve is used as the edge line in each area block.

[0015] Furthermore, the method for obtaining the defect information amount of each area block includes:

[0016] For each edge line in each area block, on the edge line, the average of the slopes between each pixel point on the edge line and its adjacent pixel points is taken as the local fluctuation of each pixel point on the edge line; the variance of the local fluctuation of all pixels on the edge line is recorded as the overall fluctuation value of the edge line;

[0017] Linear fitting is performed on the pixel points on the edge line to obtain a reference extension line of the edge line; and an extension difference value of each edge line in each area block is obtained according to the angle between each edge line in each area block and the reference extension lines of the remaining edge lines;

[0018] The defect information amount of each area block is obtained according to the overall fluctuation value and the extension difference value of all edge lines in each area block, as well as the possible indicators of the weld.

[0019] Furthermore, the method for detecting the fastening quality of the frame of the high-strength alloy template comprises:

[0020] The variance of the grayscale mean of all pixels on the edge line in each area block is used as the reflection difference of the edge line in each area block;

[0021] According to the possible indicators of the weld in each area block, the defect information amount and the reflection difference, the local bad indicators of each area block are obtained; the mean values ​​of the local bad indicators of all area blocks in the reinforcement welding image are normalized to obtain the bad fastening index; when the bad fastening index is greater than the preset defect threshold, the fastening quality of the frame of the high-strength alloy formwork is unqualified; when the bad fastening index is less than or equal to the preset defect threshold, the fastening quality of the frame of the high-strength alloy formwork is qualified.

[0022] Furthermore, the method of obtaining the overall reflectivity of the corresponding area block according to the grayscale values ​​of all pixels in each area block includes:

[0023] The mean value of the grayscale values ​​of all pixels in each area block is recorded as the overall reflectivity index of the corresponding area block.

[0024] Furthermore, the method for obtaining the defect information amount of each area block according to the overall fluctuation value and the extension difference value of all edge lines in each area block, and the possible weld index, includes:

[0025] According to the overall fluctuation value and the extension difference value of each edge line in each area block, the possible crack index of each edge line in each area block is obtained; the product of the possible weld index and the possible crack index of each area block is normalized to obtain the defect information amount of each area block.

[0026] Furthermore, the method of dividing the reinforcement welding image into different area blocks is a superpixel segmentation algorithm.

[0027] Furthermore, the method for edge detection of the reinforcement welding image is a Canny operator.

[0028] Furthermore, the method of performing curve fitting on edge pixels obtained by edge detection in each area block is a polynomial fitting method.

[0029] The present invention has the following beneficial effects:

[0030] In an embodiment of the present invention, in order to improve the accuracy of defect detection, the welding image of the reinforcement is divided into different area blocks; the weld area has better reflective ability than the template frame area, and the reinforcement is generally welded in the groove of the high-strength alloy template frame. The frame structure is prone to shadows and the shadow position is not fixed, resulting in the texture of the template frame area being more complex than the weld area. According to the degree of reflectivity and texture complexity of the area block, a measure of the possibility of the area block being a weld area, that is, a possible weld index is obtained; because there are differences in the performance of the crack edge and the weld edge, before analyzing the crack defect in the weld area, the edge line in the area block is obtained; there are differences in the degree of fluctuation of the edge, weld edge and crack edge of the template frame structure under illumination, and there are differences in the direction of the edge, weld edge and crack edge of the template frame structure, then according to the degree of fluctuation of the edge line in the area block and the different edges The difference in the extension direction of the edge line is calculated, and the defect information of the regional block is obtained in combination with the possible indicators of the weld, which is used to measure the information of the crack defect in the regional block; the amount of reflection of the crack in the imaging process is small, and the edge of the template frame and the edge of the weld are mainly caused by shadows, and their reflection amount should be greater than the reflection amount of the crack. According to the difference in the amount of reflection of the edge line in the regional block of the reinforcement welding image, the crack information in the image can be further determined, and then the frame fastening quality of the high-strength alloy template can be detected; this scheme combines the reflection degree of the regional block, the complexity of the texture, the fluctuation degree and the extension direction of the edge line in the regional block for defect detection, which effectively avoids the problem of low accuracy of crack defect detection caused by the shadow of the alloy template frame under light, thereby improving the accuracy of frame fastening detection of high-strength alloy templates. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 A flowchart of the steps of a frame fastness detection method based on high-strength alloy template structure optimization provided by one embodiment of the present invention;

[0033] Figure 2 A partial schematic diagram of a reinforcement welding image provided by an embodiment of the present invention;

[0034] Figure 3 A flowchart of a method for obtaining defect information of a region block provided by an embodiment of the present invention;

[0035] Figure 4A computer device schematic diagram of a frame tightness detection device based on high-strength alloy template structure optimization provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0036] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the frame fastness detection method based on the optimization of the high-strength alloy template structure proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0037] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0038] The following is a detailed description of a specific scheme of a frame fastness detection method based on high-strength alloy template structure optimization provided by the present invention in conjunction with the accompanying drawings.

[0039] Embodiment 1:

[0040] The present invention proposes a frame fastness detection method based on high-strength alloy template structure optimization, please refer to Figure 1 , which shows a flow chart of the steps of a frame fastness detection method based on high-strength alloy template structure optimization provided by an embodiment of the present invention, the method comprising:

[0041] Step S1: Acquire a welding image of a reinforcement member of a high-strength alloy template.

[0042] At the construction site, analyze the stability requirements of the high-strength alloy formwork under the current construction environment, determine the positions on the formwork that need to be reinforced, and weld the reinforcements to the frame of the high-strength alloy formwork at the positions that need to be reinforced. The welds of the alloy plates after reinforcement welding are relatively smooth and prone to reflection. During the image acquisition process, it is necessary to avoid as much as possible the light from being reflected by the plate and directly entering the imaging device, which would cause a large area of ​​strong reflection in the captured image. In the embodiment of the present invention, an industrial camera is installed directly in front of the reinforcement of the high-strength alloy formwork, so that the light is kept at a 45-degree angle to the plane where the reinforcement is located to illuminate the reinforcement. The industrial camera is used to capture the image of the reinforcement at the frame position of the high-strength alloy formwork to obtain the original reinforcement welding image, which is an RGB image. The original reinforcement welding image is grayscaled and denoised to obtain the reinforcement welding image. Figure 2 A partial schematic diagram of a reinforcement welding image provided by an embodiment of the present invention, such as Figure 2 As shown, Figure 2The middle white rectangular area contains a reinforcement, and the three black rectangular areas contain welds.

[0043] It should be noted that in the embodiment of the present invention, a weighted average grayscale algorithm is selected for grayscale processing, and a non-local mean filter is used for denoising processing. The specific methods are not introduced here, and they are all technical means well known to those skilled in the art. In other embodiments of the present invention, other image acquisition devices and image preprocessing algorithms can also be selected. Image acquisition and image preprocessing algorithms are both technical means well known to those skilled in the art and are not limited here.

[0044] Step S2: Divide the reinforcement welding image into different area blocks; obtain possible weld indicators of each area block according to the reflection degree and texture complexity of each area block.

[0045] Cracks near the weld affect the quality of the weld. Cracks are a source of stress concentration and can significantly reduce the fatigue strength of the welded structure or joint under alternating loads. This solution determines the frame reinforcement quality of the alloy template by analyzing the crack situation. Since cracks are usually located on the surface and near the weld, it is necessary to first determine the weld area in the image. This solution quantifies the possibility that the area block is the weld area through possible weld indicators, and then analyzes the crack defects in the weld area to obtain the defect information.

[0046] In order to determine the weld area in the image, the reinforcement welding image is first divided into different area blocks using a superpixel segmentation algorithm; in other embodiments of the present invention, the reinforcement welding image can be evenly divided into different area blocks. The superpixel segmentation algorithm is a well-known technology for those skilled in the art and will not be described in detail here.

[0047] The frame of the high-strength alloy template is a general metal part, and the frame structure will produce shadows. The weld surface is relatively smooth, so the weld area has better reflective ability than the template frame. Since the reinforcement is generally welded in the groove of the high-strength alloy template frame, the frame structure is prone to produce shadows and the shadow position is not fixed, resulting in a chaotic distribution of shadow areas and non-shadow areas in the template frame area; the weld is convex, and only the edge of the weld produces a shadow area. Compared with the texture of the weld area, the texture of the template frame area is more complex. Therefore, according to the degree of reflection and texture complexity of the area block, the possibility of the area block being a weld area, that is, the possible indicator of the weld, is analyzed.

[0048] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining possible weld indicators is: edge detection is performed on the reinforcement welding image to obtain the gradient value of each pixel in the reinforcement welding image; the variance of the gradient values ​​of all pixels in each area block is used as the texture complexity of each area block; the overall reflectivity of the corresponding area block is obtained according to the grayscale values ​​of all pixels in each area block; and the possible weld indicators of each area block are obtained according to the texture complexity and the overall reflectivity. In this embodiment, the Canny operator is used for edge detection, and the Sobel operator can also be used, which are both well-known technologies and will not be repeated here.

[0049] It should be noted that in the embodiments of the present invention, the grayscale value of the pixel point is used to present the reflectivity of the pixel point position. The greater the grayscale value, the stronger the reflectivity of the pixel point position. The mean of the grayscale values ​​of all pixels in each area block can be used as the overall reflectivity of the corresponding area block. Indicators such as mean, median and mode can all reflect the centralized distribution trend of a set of data. In other embodiments of the present invention, the mean can also be replaced by mode or median; statistical features such as standard deviation and information entropy can also be used to reflect the complexity of information.

[0050] The weld area has better reflectivity than the template frame area. The greater the overall reflectivity of the area block, the more likely it is that the area block is the weld area. The texture of the template frame area is more complex than that of the shadow and reflective parts in the weld area. The smaller the texture complexity of the area block, the more likely it is that the area block is the weld area. Therefore, the overall reflectivity is positively correlated with the possible indicators of the weld, and the texture complexity is negatively correlated with the possible indicators of the weld.

[0051] In the embodiment of the present invention, the texture complexity of each area block is negatively correlated and mapped, the product of the mapping result and the overall reflectivity is normalized, and the normalized result is used as the weld possible index of each area block. When the weld possible index of the area block is larger, the possibility that the area block is a weld area is greater; when the weld possible index of the area block is smaller, the possibility that the area block is a template frame area is greater.

[0052] In the embodiment of the present invention, the correlation between texture complexity and overall reflectivity and possible indicators of the weld can also be constructed through other basic mathematical operations, which are not limited or elaborated here.

[0053] It should be noted that the embodiment of the present invention uses an exponential function with a natural constant as the base for negative correlation mapping and uses a Sigmoid function for normalization. Other normalization methods may also be selected, such as function transformation, maximum and minimum normalization, etc., which are not limited here.

[0054] Step S3: Obtain edge lines in each area block; obtain the defect information amount of each area block according to the fluctuation degree of all edge lines in each area block and the difference between the extension directions of different edge lines, as well as possible indicators of the weld.

[0055] The cracks in the weld area affect the reinforcement quality of the weld to the reinforcement and the alloy template. The crack edge is different from the normal welding texture, that is, the weld edge. Therefore, before analyzing the crack defects in the weld area, it is necessary to obtain the edge line in the area block.

[0056] In the embodiment of the present invention, a curve fitting is performed on the edge pixels obtained by edge detection in each area block, and the fitted curve is used as the edge line in each area block. The edge line may represent the edge of a weld, the edge of a crack, and the edge of a template frame structure.

[0057] It should be noted that, in order to retain the fluctuation characteristics of edge pixels as much as possible, a polynomial fitting method with a complex curve shape is selected in the embodiment of the present invention to perform curve fitting on edge pixels to obtain edge lines within the region block. The flexibility of fitting is controlled by selecting an appropriate polynomial order. In the embodiment of the present invention, the polynomial order for fitting edge pixels is 4, and the implementer can set it according to the specific situation.

[0058] It should be noted that before fitting the edge pixels, it is necessary to determine whether there are other edge pixels in the eight-neighborhood of each edge pixel. If not, it means that the edge pixel is a noise point generated during the image acquisition process and does not need to participate in curve fitting; if so, it means that the edge pixel is not a noise point and these edge pixels are the objects of curve fitting.

[0059] The template frame structure is relatively regular, such as Figure 2 As shown in the figure, the edge of the template frame structure is mainly composed of the frame shadow and its concave and convex lines, and these edges are relatively regular; the edge fluctuation of the weld depends on many factors, and usually the fluctuation of the weld edge is slight; the crack is formed by the fracture of the material or stress concentration, and the crack edge usually has a more irregular and sharp shape. Compared with the weld edge, the crack edge usually shows more obvious fluctuation or irregularity. Therefore, the fluctuation degree of the edge of the template frame structure, the weld edge and the crack edge becomes stronger in turn.

[0060] The welding process of high-strength alloy templates is prone to longitudinal cracks. The longitudinal cracks run along the direction of the weld. Other types of cracks are not as common as transverse cracks. This scheme only analyzes the longitudinal cracks of the weld. The reinforcement is in the shape of a long strip. When the reinforcement is welded to the high-strength alloy frame, the welding direction of the weld is along the direction of the frame; Figure 2 As shown, Figure 2The edge of the template frame structure is close to the horizontal direction, the longitudinal or transverse direction of the crack is relative to the weld, the weld direction in the two black rectangular areas on the left is horizontal and parallel, and the weld direction in the black rectangular area on the right is horizontal and vertical. In the embodiment of the present invention, the extension direction of the edge line represents the overall direction of the edge pixel points of the edge line fitting, and Figure 2 Most of the welds are horizontal, so the extension direction of most edge lines in each area block is close to the horizontal direction; the extension direction of a small part of the edge lines is close to the vertical direction, and these edge lines are more likely to be crack edges. The greater the difference between the extension direction of each edge line in the area block and the other edge lines, the greater the possibility that the edge line is a crack edge, and the more crack defect information contained in the area block.

[0061] Cracks are usually located on the surface and near the weld. The larger the weld possible index of the area block, the greater the possibility that the area block is a weld area, and the greater the possibility that the area block contains crack information.

[0062] In summary, the amount of information about the crack defect in the regional block, i.e., the defect information amount, is quantified based on the fluctuation degree of the edge line in the regional block, the difference between the extension directions of different edge lines, and the possible indicators of the weld.

[0063] See also Figure 3 , which shows a flowchart of a method for obtaining defect information of a regional block provided by an embodiment of the present invention, the method comprising:

[0064] Step S310: For each edge line in each area block, on the edge line, the average of the slopes between each pixel point on the edge line and its adjacent pixel points is taken as the local fluctuation of each pixel point on the edge line; the variance of the local fluctuations of all pixels on the edge line is recorded as the overall fluctuation value of the edge line.

[0065] It is known that the fluctuation degrees of the template frame structure, weld edge and crack edge become stronger in sequence. This embodiment presents the local fluctuation of each pixel on the edge line through the slope between each pixel on the edge line and its adjacent pixel points. The more obvious the local fluctuation of the pixel on the edge line, the greater the local fluctuation of the pixel. The local fluctuation of all pixels on the edge line is combined to obtain the overall fluctuation of the edge line. The larger the overall fluctuation value, the more obvious the fluctuation of the edge line, and the greater the possibility that the edge line is a crack edge.

[0066] It should be noted that in other embodiments of the present invention, the variance may be replaced by statistics such as standard deviation and information entropy to reflect fluctuations.

[0067] Step S320: linearly fit the pixel points on the edge line to obtain the baseline extension line of the edge line; obtain the extension difference value of each edge line in each area block according to the angle between each edge line in each area block and the baseline extension lines of the other edge lines.

[0068] In an embodiment of the present invention, Hough straight line fitting is performed on the pixel points on the edge line to obtain the reference extension line of the edge line; in other embodiments of the present invention, the principal component analysis algorithm can also be used to perform linear fitting on the pixel points on the edge line, and the principal axis obtained by fitting is used as the reference extension line of the edge line. It should be noted that the principal axis obtained by fitting the principal component analysis algorithm is a straight line.

[0069] Among them, Hough straight line fitting and principal component analysis algorithms are well-known technologies to those skilled in the art and will not be described in detail here.

[0070] It should be noted that the direction of the fitting line obtained by linear fitting of the pixels on the edge line is usually consistent with the overall distribution direction or trend of the pixels on the edge line, so the direction of the baseline extension line of the edge line can represent the overall trend direction of the pixels on the edge line.

[0071] Preferably, in some possible implementations of the embodiments of the present invention, the extension difference value is obtained by: for each area block, the average of the sine values ​​of the angles between each edge line in the area block and the other edge lines is recorded as the extension difference value of each edge line in the area block. The larger the extension difference value of the edge line is, the greater the possibility that the edge line is a crack edge.

[0072] In other embodiments of the present invention, the average value of the angles between each edge line in the region block and the remaining edges may be directly recorded as the extension difference value of each edge line in the region block.

[0073] Step S330: Obtain the defect information amount of each area block according to the overall fluctuation value and extension difference value of all edge lines in each area block, as well as the possible indicators of the weld.

[0074] Preferably, the method for obtaining the defect information includes: obtaining the possible crack index of each edge line in each area block according to the overall fluctuation value and the extension difference value of each edge line in each area block; normalizing the product of the possible weld index and the possible crack index of each area block to obtain the defect information of each area block. If the overall fluctuation value of the edge line is larger, the possibility that the edge line is a crack edge is greater; if the extension difference value of the edge line is larger, the possibility that the edge line is a crack edge is greater, then the overall fluctuation value and the extension difference value are both positively correlated with the possible crack index.

[0075] In a specific implementation of the embodiment of the present invention, the specific formula of the defect information amount is expressed as:

[0076]

[0077] In the formula, is the defect information amount of each area block; F is the possible weld index of each area block; J is the total number of edge lines in each area block; is the overall volatility of the jth edge line in each area block; is the extension difference value of the jth edge line in each area block; It is a preset positive number, with an empirical value of 1, which is used to prevent A value of 0 results in meaningless defect information; is the possible crack indicator of the jth edge line in each area block; sin is the sine function.

[0078] It should be noted that when When the value is larger, it means that the edge line in each area block is more likely to be a crack edge, and the area block contains more crack defect information. The larger the weld index F of the area block, the greater the possibility that the area block is a weld area, because cracks are usually located on the weld surface and near the weld. The more crack information the area block contains, the greater the defect information amount. The bigger.

[0079] Step S4: The frame fastening quality of the high-strength alloy template is inspected based on the difference in the amount of reflection of the edge lines in all area blocks of the reinforcement welding image, possible indicators of the weld and the amount of defect information.

[0080] The crack defect in the alloy template welding process is mainly caused by partial fracture inside the weld. The reflection amount of the crack during the imaging process is small. The edge of the template frame and the edge of the weld are mainly caused by shadows. In this scheme, the reflection amount is presented by grayscale value, so the grayscale of the crack edge is smaller than the grayscale of the edge of the template frame and the edge of the weld, and the grayscale of the edge of the template frame and the edge of the weld is close. If the difference in the reflection amount of the edge line in the area block is small, the possibility of cracks in the area block is small; if the difference in the reflection amount of the edge line in the area block is large, the possibility of cracks in the area block is large, and the fastening quality of the area block may be worse. At the same time, the defect information of the area block and the possible indicators of the weld are analyzed to effectively improve the accuracy of the fastening quality detection of the frame of the high-strength alloy template.

[0081] Preferably, the method for obtaining the poor fastening index is: taking the variance of the grayscale mean of all pixel points on the edge lines in each area block as the reflection difference of the edge lines in each area block; obtaining the local poor index of each area block according to the possible weld index, defect information amount and reflection difference of each area block; normalizing the mean of the local poor index of all area blocks in the reinforcement welding image to obtain the poor fastening index.

[0082] In a specific implementation of the embodiment of the present invention, the specific formula of the poor fastening index is expressed as:

[0083]

[0084] Where, P is the poor fastening index; U is the total number of regional blocks in the reinforcement welding image; is the possible weld index of the u-th area block in the reinforcement welding image; is the defect information amount of the u-th area block in the reinforcement welding image; The reflection difference of the u-th area block in the reinforcement welding image; is the local defect index of the u-th area block in the reinforcement welding image; Norm is the normalization function.

[0085] It should be noted that when the weld may indicate When the defect information amount is larger, the possibility that the u-th area block in the reinforcement welding image is the weld area is greater, because the crack is usually located on the weld surface and nearby, so the fastening quality of this area block may be worse; When the value is larger, the more crack information is contained in the u-th area block, the worse the fastening quality of the area block may be, and the larger the fastening failure index is. When the value is larger, the possibility of cracks existing in the u-th area block is greater, the fastening quality of the area block may be worse, and the fastening defect index is larger.

[0086] Therefore, the possible indicators of the weld, the amount of defect information and the reflection difference are all positively correlated with the local bad indicators. In the embodiment of the present invention, the correlation between the possible indicators of the weld, the amount of defect information and the reflection difference and the local bad indicators can also be constructed through other basic mathematical operations, which are not limited or elaborated here.

[0087] When the poor fastening index is greater than the preset defect threshold, there are cracks at the weld position between the high-strength alloy formwork and the reinforcement, and the frame reinforcement quality of the high-strength alloy formwork is unqualified; when the poor fastening index is less than or equal to the preset defect threshold, there are no cracks at the weld position between the high-strength alloy formwork and the reinforcement, and the frame reinforcement quality of the high-strength alloy formwork is qualified.

[0088] It should be noted that the preset defect threshold in the embodiment of the present invention is an empirical value of 0.45, and the implementer can set it according to the specific situation.

[0089] At the construction site, in order to ensure that the frame of the high-strength alloy formwork has sufficient strength and stability to withstand the pressure generated during the concrete pouring process, after the formwork frame is reinforced and welded, this solution is used to test the welding quality of the frame of the high-strength alloy formwork and the reinforcement parts, which can accurately detect the strength and integrity of the weld and ensure the firmness and stability of the weld. This measure not only significantly reduces the risk of excessive waterline on the wall after pouring concrete, but also effectively improves the overall visual effect and construction quality, making the wall straight and the joints delicate after pouring, greatly improving the overall look of the wall, giving the building a higher sense of quality and professionalism, but also extending the service life of the high-strength alloy formwork and reducing the workload of later repairs.

[0090] So far, the present invention is completed.

[0091] In summary, in an embodiment of the present invention, the possible weld index of the corresponding area block is obtained according to the degree of reflection and the complexity of the texture of each area block of the reinforcement welding image, and the amount of defect information of each area block is obtained in combination with the difference between the degree of fluctuation of the edge line in each area block and the extension direction of different edge lines; the frame reinforcement quality of the high-strength alloy formwork is detected according to the difference in the amount of reflection of the edge line in the area block of the reinforcement welding image, the possible weld index and the amount of defect information. The present invention performs defect detection based on multiple factors such as the degree of reflection of the area block, the complexity of the texture, and the degree of fluctuation and the extension direction of the edge line in the area block, thereby improving the accuracy of the frame fastening detection of the high-strength alloy formwork.

[0092] Embodiment 2:

[0093] Figure 4 A computer device schematic diagram of a frame fastness detection device based on high-strength alloy template structure optimization provided by an embodiment of the present invention. For example, Figure 4 As shown, the computer device includes: a memory 501, a processor 502, and a computer program 503 stored in the memory 501 and running on the processor 502, wherein when the processor 502 executes the computer program 503, the computer device can execute any of the frame tightness detection methods based on high-strength alloy template structure optimization introduced above.

[0094] In addition, an embodiment of the present application also protects a device, which may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to execute a border tightness detection method based on high-strength alloy template structure optimization provided in an embodiment of the present application.

[0095] In this embodiment, the functional modules of the device can be divided according to the above method example. For example, each functional module can be corresponded, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0096] It should be understood that the device provided in this embodiment is used to execute the above-mentioned frame tightness detection method based on high-strength alloy template structure optimization, and thus can achieve the same effect as the above-mentioned implementation method.

[0097] In the case of an integrated unit, the device may include a processing module and a storage module. When the device is applied to a device, the processing module may be used to control and manage the actions of the device. The storage module may be used to support the device to execute mutual program codes, etc.

[0098] The processing module may be a processor or a controller, which may implement or execute various exemplary logic blocks, modules and circuits disclosed in the present application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc. The storage module may be a memory.

[0099] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0100] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0101] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A frame fastness detection method based on high-strength alloy template structure optimization, characterized in that: The method includes: Obtain welding images of reinforcements of high-strength alloy templates; Divide the reinforcement welding image into different area blocks; obtain the possible weld index of each area block according to the reflection degree and texture complexity of each area block; Obtain edge lines in each area block; obtain defect information of each area block according to the fluctuation degree of all edge lines in each area block and the difference between the extension directions of different edge lines, as well as the possible indicators of the weld; The frame fastening quality of the high-strength alloy template is detected based on the difference in the amount of reflection of the edge lines in all area blocks of the reinforcement welding image, the possible indicators of the weld and the amount of defect information; The method for obtaining possible weld indicators of each area block includes: Perform edge detection on the reinforcement welding image to obtain the gradient value of each pixel in the reinforcement welding image; The variance of the gradient values ​​of all pixels in each area block is used as the texture complexity of each area block; the overall reflectivity of the corresponding area block is obtained according to the grayscale values ​​of all pixels in each area block; and the possible weld index of each area block is obtained according to the texture complexity and the overall reflectivity; The method for obtaining the defect information amount of each area block includes: For each edge line in each area block, on the edge line, the average of the slopes between each pixel point on the edge line and its adjacent pixel points is taken as the local fluctuation of each pixel point on the edge line; the variance of the local fluctuation of all pixels on the edge line is recorded as the overall fluctuation value of the edge line; Linear fitting is performed on the pixel points on the edge line to obtain a reference extension line of the edge line; and an extension difference value of each edge line in each area block is obtained according to the angle between each edge line in each area block and the reference extension lines of the remaining edge lines; The defect information amount of each area block is obtained according to the overall fluctuation value and the extension difference value of all edge lines in each area block, as well as the possible indicators of the weld.

2. A frame fastness detection method based on high-strength alloy template structure optimization according to claim 1, characterized in that: The method for obtaining the edge line in each area block includes: Curve fitting is performed on the edge pixels obtained by edge detection in each area block, and the fitted curve is used as the edge line in each area block.

3. The frame fastness detection method based on high-strength alloy template structure optimization according to claim 1 is characterized in that: The method for detecting the fastening quality of the frame of the high-strength alloy template comprises: The variance of the grayscale mean of all pixels on the edge line in each area block is used as the reflection difference of the edge line in each area block; According to the possible indicators of the weld in each area block, the defect information amount and the reflection difference, the local bad indicators of each area block are obtained; the mean values ​​of the local bad indicators of all area blocks in the reinforcement welding image are normalized to obtain the bad fastening index; when the bad fastening index is greater than the preset defect threshold, the fastening quality of the frame of the high-strength alloy formwork is unqualified; when the bad fastening index is less than or equal to the preset defect threshold, the fastening quality of the frame of the high-strength alloy formwork is qualified.

4. A frame fastness detection method based on high-strength alloy template structure optimization according to claim 1, characterized in that: The method for obtaining the overall reflectivity of the corresponding area block according to the grayscale values ​​of all pixels in each area block includes: The mean value of the grayscale values ​​of all pixels in each area block is recorded as the overall reflectivity index of the corresponding area block.

5. The frame fastness detection method based on high-strength alloy template structure optimization according to claim 1 is characterized in that: The method for obtaining the defect information amount of each area block according to the overall fluctuation value and the extension difference value of all edge lines in each area block, and the possible weld index, comprises: According to the overall fluctuation value and the extension difference value of each edge line in each area block, the possible crack index of each edge line in each area block is obtained; the product of the possible weld index and the possible crack index of each area block is normalized to obtain the defect information amount of each area block.

6. The frame fastness detection method based on high-strength alloy template structure optimization according to claim 1 is characterized in that: The method for dividing the reinforcement welding image into different area blocks is a superpixel segmentation algorithm.

7. A frame fastness detection method based on high-strength alloy template structure optimization according to claim 1, characterized in that: The method for edge detection of the reinforcement welding image is the Canny operator.

8. The frame fastness detection method based on high-strength alloy template structure optimization according to claim 2 is characterized in that: The method for performing curve fitting on edge pixel points obtained by edge detection in each area block is a polynomial fitting method.

Images