Method for evaluating roughness of composite surface of concrete precast member based on two-dimensional image

Abstract

The application provides a concrete prefabricated component superimposed surface roughness evaluation method based on a two-dimensional image, and the method comprises the following steps: performing image gray processing on the collected two-dimensional image to obtain a gray image of the concrete prefabricated component superimposed surface; calculating the image gray co-occurrence matrix under different image gray levels; calculating the contrast and gray entropy of the image gray co-occurrence matrix under different image gray levels, and calculating the weight corresponding to the contrast and gray entropy; performing weighted summation on the contrast and gray entropy of the gray image of the concrete prefabricated component superimposed surface under the maximum gray level, and calculating the roughness index of the concrete prefabricated component superimposed surface, so as to evaluate the roughness state of the concrete prefabricated component superimposed surface according to the roughness index. The application has the advantages of realizing quantitative evaluation of the roughness of the concrete prefabricated component superimposed surface and convenient operation.

Description

Technical Field

[0001] This invention relates to the field of concrete construction, and more particularly to a method for evaluating the roughness of the overlapping surfaces of precast concrete components based on two-dimensional images. Background Technology

[0002] In the construction industry, prefabricated concrete structures are widely used in various building projects. The connection quality between prefabricated concrete components plays a decisive role in the performance of the entire structure, and the roughness of the mating surface is one of the key factors affecting the connection effect. Appropriate mating surface roughness can significantly enhance the bond between new and old concrete, effectively improving the structure's integrity, seismic performance, and durability. Therefore, mating surface roughness has a significant impact on the performance of prefabricated concrete structures. Accurate evaluation of mating surface roughness can ensure that the mating surface roughness of prefabricated components from different production batches is within the reasonable range required by design, guaranteeing the consistency of the mechanical properties of the connections between components during on-site assembly, and helping to fully leverage the mechanical performance advantages of prefabricated concrete structures.

[0003] Evaluation methods for the roughness of the composite surface of precast concrete components include the stylus method and the template comparison method. However, the stylus method suffers from drawbacks such as damage to the component surface during testing, low testing efficiency, and difficulty in comprehensively reflecting the overall roughness of the composite surface. The template comparison method is highly subjective, with different operators using different judgment standards, leading to inconsistent and inaccurate evaluation results. With the increasing demands for the quality of precast concrete structures, existing evaluation methods are no longer sufficient to meet the needs of practical engineering projects for accurate, efficient, and non-destructive testing of the roughness of composite surfaces.

[0004] Furthermore, due to the diverse sizes and shapes of precast concrete components, the overlapping surfaces of large-sized components present challenges in inspection. Simultaneously, the overlapping surfaces of precast concrete components are typically large, exhibiting significant unevenness in roughness, meaning there may be large differences in roughness between different local areas, making overall evaluation difficult. For example, for large-sized components, one area might have good roughness while another has poor roughness. Existing roughness testing methods can only describe local roughness, making it difficult to simultaneously consider both overall and local roughness. For small-sized components, size effects may render conventional evaluation methods inapplicable. For instance, for thin-walled components, minute surface changes have a significant impact on roughness, and traditional testing methods cannot accurately measure these minute changes, leading to large errors in roughness assessment. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a method for evaluating the roughness of the composite surface of precast concrete components based on two-dimensional images, which is convenient to operate and has high evaluation accuracy.

[0006] To solve the above-mentioned technical problems, the technical solution proposed by this invention is as follows:

[0007] A method for evaluating the surface roughness of composite surfaces of precast concrete components based on two-dimensional images includes the following steps:

[0008] Step 1), acquire a two-dimensional image of the overlapping surface of the precast concrete components, and perform image grayscale processing on the acquired two-dimensional image to obtain a grayscale image of the overlapping surface of the precast concrete components.

[0009] Step 2), calculate the image gray-level co-occurrence matrix of the gray-level image of the composite surface of the precast concrete component under different image gray-level conditions; based on the calculated image gray-level co-occurrence matrices, calculate the contrast and gray-level entropy of the image gray-level co-occurrence matrix under different image gray-level conditions;

[0010] Step 3), calculate the weights corresponding to contrast and gray entropy of the image gray-level co-occurrence matrix under different image gray-level conditions respectively;

[0011] Step 4) Using the calculated contrast weight and gray entropy weight, the contrast and gray entropy of the grayscale image of the composite surface of the precast concrete component are weighted and summed at the maximum gray level to calculate the roughness index of the composite surface of the precast concrete component, so as to evaluate the roughness state of the composite surface of the precast concrete component according to the roughness index.

[0012] As a further improvement to the above technical solution:

[0013] In step 2), the formula for calculating each element in the image gray-level co-occurrence matrix of the composite surface of the precast concrete component under different image gray-level conditions is as follows:

[0014]

[0015] in, To represent the grayscale image of the overlapping surface of precast concrete components, the upper grayscale value is the value of the vertically adjacent pixels. The grayscale below is The probability of; This represents the total number of pixels in the grayscale image of the overlapping surfaces of precast concrete components. Indicates a grayscale value of Starting from the pixel, according to a distance of 1 and a direction of 0 degrees, the grayscale value is... The number of pixel pairs; This represents the number of gray levels in a grayscale image under different grayscale conditions.

[0016] As a further improvement to the above technical solution:

[0017] In step 2), based on the calculated gray-level co-occurrence matrices of each image, the following formula is used to calculate the gray-level co-occurrence matrix of the first image. The contrast and gray-level entropy of the image gray-level co-occurrence matrix under the condition of each image gray level:

[0018]

[0019]

[0020] in, In the first Contrast of the gray-level co-occurrence matrix of an image under a given image gray level condition; In the first Gray-level entropy of the gray-level co-occurrence matrix of an image under gray-level conditions. , This represents the total number of image grayscale level conditions;

[0021] Calculate under different image gray level conditions The contrast matrix is ​​obtained by analyzing the contrast of the gray-level co-occurrence matrix of the image. And calculation under different image gray level conditions The gray-level entropy matrix is ​​obtained by taking the gray-level co-occurrence matrix of the image. .

[0022] In step 3), the weights of contrast and grayscale entropy are calculated using the entropy weight method, including:

[0023] The image gray-level co-occurrence matrix is ​​used under different image gray-level conditions. Contrast matrix below and grayscale entropy matrix After normalization, the normalized contrast matrix and gray-level entropy matrix are obtained:

[0024]

[0025]

[0026]

[0027]

[0028] in, This is the normalized contrast matrix. The normalized gray-level entropy matrix, , They are respectively 、 Elements in the matrix In the first Contrast of the gray-level co-occurrence matrix of an image under a given number of gray levels; In the first Gray-level entropy of the gray-level co-occurrence matrix of an image under a given gray level condition;

[0029] The information entropy values ​​of contrast and grayscale entropy are calculated using the following formulas based on the normalized contrast matrix and grayscale entropy matrix:

[0030]

[0031]

[0032] in, The information entropy value of contrast. This represents the information entropy value of the grayscale entropy.

[0033] The weights of contrast and grayscale entropy are determined based on the information entropy values ​​of contrast and grayscale entropy.

[0034] Based on the information entropy values ​​of contrast and grayscale, the weights of contrast and grayscale are calculated using the following formulas:

[0035]

[0036]

[0037] in, As the weight of contrast, The weight is the grayscale entropy.

[0038] In step 4), the contrast and gray-level entropy of the gray-level image of the composite surface of the precast concrete component are weighted and summed at the maximum gray level using the following formula, and the roughness index of the composite surface of the precast concrete component is calculated:

[0039]

[0040] in, It is a roughness index for the overlapping surfaces of precast concrete components. As the weight of contrast, The weights are the grayscale entropy. To achieve the maximum gray level Contrast after normalization under certain conditions To achieve the maximum gray level The gray entropy after normalization under certain conditions.

[0041] In step 2), before calculating the image gray-level co-occurrence matrix of the gray-level image of the composite surface of the precast concrete component under different image gray-level conditions, the method further includes setting image gray-level conditions, which include at least three conditions, with the maximum image gray-level being 256 levels.

[0042] In step 1), before performing image grayscale processing on the acquired two-dimensional image, the method further includes: removing background impurities from the acquired two-dimensional image and retaining the main body of the composite surface of the precast concrete component to obtain a two-dimensional image after removing impurities; performing image grayscale processing on the two-dimensional image after removing impurities using a threshold segmentation method to form a grayscale image of the composite surface of the precast concrete component; after performing image grayscale processing, the method further includes performing image denoising processing on the grayscale image of the composite surface of the precast concrete component using a Gaussian filtering method.

[0043] A computer device includes a processor and a memory for storing a computer program, characterized in that the processor is configured to execute the computer program to perform the method as described above.

[0044] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method described above.

[0045] Compared with the prior art, the advantages of the present invention are as follows:

[0046] This invention utilizes in-depth analysis of the grayscale features of two-dimensional images of the composite surface. Based on the grayscale co-occurrence matrix (GCMM) calculated at different grayscale levels, it calculates the contrast and grayscale entropy of the GCMM under different image grayscale conditions. Then, it uses the contrast and grayscale entropy of the GCMM under different image grayscale conditions to calculate the weights corresponding to contrast and grayscale entropy. Finally, it uses these weights to perform a weighted summation of the contrast and grayscale entropy at the maximum grayscale level to obtain a quantitative evaluation index of the composite surface roughness. This constructs a quantitative model directly related to the composite surface roughness, which comprehensively considers both contrast and grayscale entropy parameters and dynamically adjusts the corresponding weights. This allows for the simultaneous consideration of the overall and local roughness of the component, achieving a non-destructive and efficient quantitative evaluation of the roughness of the composite surface of precast concrete components. Furthermore, the roughness index obtained by this invention enables a fine quantitative evaluation of the roughness of the composite surface of precast concrete components, and also allows for a quantitative comparison of the roughness of different components, providing a more intuitive and accurate comparison to determine the roughness of different components.

[0047] Meanwhile, this invention can adaptively adjust the weights of contrast and grayscale entropy in the grayscale image of the composite surface of precast concrete components according to their actual state. Based on the contrast and grayscale entropy at different grayscale levels, the weights determined by the information entropy theory are objective weights. This can accurately determine the objective weights of contrast and grayscale entropy in real time under different working conditions, avoiding the problem of inaccurate results caused by subjective judgment. By using the weights determined in this way to perform a weighted summation of the contrast and grayscale entropy at the maximum grayscale level, the roughness index of the composite surface of precast concrete components can be accurately calculated, which greatly improves the accuracy of the detection results and thus achieves accurate and reliable evaluation of the roughness of the composite surface.

[0048] Finally, this invention can achieve accurate assessment of the roughness of the composite surface by relying solely on image analysis, which can effectively reduce the problems of human interference and poor detection accuracy, and does not require destructive operations on the component surface. Thus, it provides reliable data support for the quality control and performance optimization of concrete structures, and ensures the reliable and effective assembly of concrete structures. Attached Figure Description

[0049] The invention will now be described in more detail with reference to embodiments and the accompanying drawings.

[0050] Figure 1 This is a flowchart of the method for evaluating the roughness of the composite surface of precast concrete components based on two-dimensional images, according to the present invention.

[0051] Figure 2 This is a schematic diagram of the original images of the overlapping surfaces of the eaves boards collected in an embodiment of the present invention, wherein (a) corresponds to the original images of the first group of overlapping surfaces of the eaves boards collected, and (b) corresponds to the original images of the second group of overlapping surfaces of the eaves boards collected.

[0052] Figure 3 This is a schematic diagram of the overlapping surface of the eaves panels obtained after image optimization processing in an embodiment of the present invention. (a) corresponds to the image obtained after image optimization processing of the original image of the overlapping surface of the eaves panels collected in the first group, and (b) corresponds to the image obtained after image optimization processing of the original image of the overlapping surface of the eaves panels collected in the second group. Detailed Implementation

[0053] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments, but this does not limit the scope of protection of the present invention.

[0054] The surface roughness of precast concrete components is influenced by various factors, exhibiting complexity and diversity. For larger components, the surface roughness is typically large, resulting in significant non-uniformity. For smaller components, traditional inspection methods cannot accurately measure minute surface variations. Contrast ratio directly reflects the grayscale differences in different areas of the surface, allowing for quick assessment of surface undulations. Its calculation is relatively simple, providing a rapid parameter reflecting roughness. Grayscale entropy, on the other hand, considers the statistical characteristics of the overall image's grayscale distribution, comprehensively reflecting the complexity of the surface's microstructure. It is sensitive to subtle changes in surface features, allowing for the detection of even minute roughness variations. However, a single parameter is insufficient to comprehensively and accurately characterize the overall roughness of the precast concrete component's surface roughness. For example, in cases with localized protrusions but a relatively uniform overall grayscale distribution, contrast ratio or grayscale entropy alone may not accurately characterize the overall roughness. Combining both contrast ratio and grayscale entropy provides a more accurate assessment.

[0055] This invention uses a combination of contrast and gray-level entropy to characterize the surface roughness of precast concrete components. Contrast focuses on the gray-level differences between adjacent areas, highlighting local undulations, while gray-level entropy focuses on the overall gray-level distribution disorder, reflecting the complexity of the microstructure. Thus, by combining the two, the surface roughness can be described more comprehensively from both local and global perspectives. This allows for the integration of more roughness-related information, more accurately reflecting the true surface roughness, effectively overcoming the limitations of a single parameter, reducing errors caused by the one-sidedness of a single indicator, and improving the reliability and accuracy of the assessment.

[0056] Meanwhile, considering that the higher the gray level, the richer the detail information that can be obtained from image processing, and that the contrast and gray entropy obtained under different gray level conditions can represent the contrast and gray entropy of the overlapping surface under different levels of detail information richness, this invention determines the objective weights of contrast and gray entropy based on the contrast and gray entropy under different gray levels using the information entropy theory. This can accurately represent the importance of contrast and gray entropy under objective conditions, avoiding the problem of inaccurate results caused by subjective judgment. By using the weights determined in this way to perform a weighted summation of the contrast and gray entropy at the maximum gray level, the roughness index of the overlapping surface of precast concrete components can be accurately calculated.

[0057] Traditional methods, such as those using standard specifications or other evaluation techniques, can only roughly determine whether roughness is acceptable or unacceptable. For example, they typically categorize roughness into very rough, moderately rough, non-rough, or acceptable / unacceptable levels, failing to quantify the specific degree of roughness and making it difficult to compare the roughness of different components. While methods like the stylus method can obtain some roughness index, they only provide a descriptive value without defined boundaries, still unable to effectively quantify the specific degree of roughness. The method of this invention provides a roughness index with defined boundaries. This index not only comprehensively considers the overall and local roughness but also allows for precise quantification of roughness. Based on this precise quantification, it facilitates a concrete comparison of the roughness of different components. Furthermore, this invention achieves accurate assessment of the roughness of composite surfaces solely through image analysis, effectively reducing human interference and poor detection accuracy. It also eliminates the need for destructive operations on the component surface, thus providing reliable data support for the quality control and performance optimization of concrete structures, ensuring reliable and effective assembly of concrete structures.

[0058] Figure 1 An embodiment of the method for evaluating the surface roughness of precast concrete components based on two-dimensional images is shown, which includes the following steps:

[0059] Step 1), acquire a two-dimensional image of the overlapping surface of the precast concrete components, and perform image grayscale processing on the acquired two-dimensional image to obtain a grayscale image of the overlapping surface of the precast concrete components.

[0060] Step 2), calculate the gray-level co-occurrence matrix of the gray-level image of the composite surface of the precast concrete component under different image gray-level conditions; based on the calculated gray-level co-occurrence matrix of each image, calculate the contrast and gray-level entropy of the image gray-level co-occurrence matrix under different image gray-level conditions;

[0061] Step 3), calculate the weights corresponding to contrast and gray entropy of the image gray-level co-occurrence matrix under different image gray-level conditions respectively;

[0062] Step 4) Using the calculated contrast weight and gray entropy weight, the contrast and gray entropy of the grayscale image of the composite surface of the precast concrete component are weighted and summed at the maximum gray level to calculate the roughness index of the composite surface of the precast concrete component, so as to evaluate the roughness state of the composite surface of the precast concrete component based on the roughness index.

[0063] This invention utilizes in-depth analysis of the grayscale features of two-dimensional images of the composite surface. Based on the grayscale co-occurrence matrix (GCMM) calculated at different grayscale levels, it calculates the contrast and grayscale entropy of the GCMM under different image grayscale conditions. Then, it uses the contrast and grayscale entropy of the GCMM under different image grayscale conditions to calculate the weights corresponding to contrast and grayscale entropy. Finally, it uses these weights to perform a weighted summation of the contrast and grayscale entropy at the maximum grayscale level to obtain a quantitative evaluation index of the composite surface roughness. This constructs a quantitative model directly related to the composite surface roughness, which comprehensively considers both contrast and grayscale entropy parameters and dynamically adjusts the corresponding weights. This allows for the simultaneous consideration of the overall and local roughness of the component, achieving a non-destructive and efficient quantitative evaluation of the roughness of the composite surface of precast concrete components. Furthermore, the roughness index obtained by this invention enables a fine quantitative evaluation of the roughness of the composite surface of precast concrete components, and also allows for a quantitative comparison of the roughness of different components, providing a more intuitive and accurate comparison to determine the roughness of different components.

[0064] Meanwhile, this invention can adaptively adjust the weights of contrast and grayscale entropy in the grayscale image of the composite surface of precast concrete components according to their actual state. Based on the contrast and grayscale entropy at different grayscale levels, the weights determined by the information entropy theory are objective weights. This can accurately determine the objective weights of contrast and grayscale entropy in real time under different working conditions, avoiding the problem of inaccurate results caused by subjective judgment. By using the weights determined in this way to perform a weighted summation of the contrast and grayscale entropy at the maximum grayscale level, the roughness index of the composite surface of precast concrete components can be accurately calculated, which greatly improves the accuracy of the detection results and thus achieves accurate and reliable evaluation of the roughness of the composite surface.

[0065] Further, in step 1), before performing image grayscale processing on the acquired two-dimensional image, the method further includes: removing the background impurities in the acquired two-dimensional image and retaining the main body of the composite surface of the precast concrete component to obtain a two-dimensional image after removing impurities; performing image grayscale processing on the two-dimensional image after removing impurities using a threshold segmentation method to form a grayscale image of the composite surface of the precast concrete component; and after performing image grayscale processing, performing image denoising processing on the grayscale image of the composite surface of the precast concrete component using a Gaussian filtering method.

[0066] This provides clear and accurate two-dimensional image grayscale information of the composite surface of precast concrete components before evaluation, ensuring the accuracy of the subsequent surface roughness index of the precast concrete components.

[0067] More preferably, in step 1), the two-dimensional image of the overlapping surface of the precast concrete components has more than 20 million pixels to provide a high-definition two-dimensional image of the overlapping surface, ensuring the reliability and accuracy of subsequent analysis data.

[0068] Further, in step 2), the formula for calculating each element in the gray-level co-occurrence matrix of the composite surface of the precast concrete component under different image gray-level conditions is as follows:

[0069] (1)

[0070] in, To represent the grayscale image of the overlapping surface of precast concrete components, the upper grayscale value is the value of the vertically adjacent pixels. The grayscale below is The probability of; This represents the total number of pixels in the grayscale image of the overlapping surfaces of precast concrete components. Indicates a grayscale value of Starting from the pixel, according to a distance of 1 and a direction of 0 degrees, the grayscale value is... The number of pixel pairs; This refers to the number of gray levels in a grayscale image under different grayscale conditions. (Grayscale level) Different values ​​can be set as needed to form different image grayscale conditions, such as 50, 100, 200, etc., to achieve image processing of different grayscale levels.

[0071] In this embodiment, the total number of pixels and the number of pixel pairs in the grayscale image of the composite surface of the precast concrete components can be automatically identified by the grayscale co-occurrence matrix function. The present invention uses the above formula (1) to quickly and accurately calculate the image grayscale co-occurrence matrix under different image grayscale conditions, thereby providing a basis for subsequently obtaining the contrast and grayscale entropy of the image grayscale co-occurrence matrix under different image grayscale conditions.

[0072] Further, in step 2), based on the calculated gray-level co-occurrence matrices of each image, the following formula is used to calculate the gray-level co-occurrence matrix of the first image. Contrast and gray entropy of the gray-level co-occurrence matrix of an image under given image gray levels:

[0073] (2)

[0074] (3)

[0075] in, In the first Contrast of the gray-level co-occurrence matrix of an image under a given image gray level condition; In the first Gray-level entropy of the gray-level co-occurrence matrix of an image under gray-level conditions. , This represents the total number of image grayscale conditions. For the current number The number of gray levels in a grayscale image given a given number of gray levels.

[0076] Calculate under different image gray level conditions The contrast matrix is ​​obtained by analyzing the contrast of the gray-level co-occurrence matrix of the image. And calculation under different image gray level conditions The gray-level entropy matrix is ​​obtained by taking the gray-level co-occurrence matrix of the image. .

[0077] For example, set seven image grayscale conditions respectively. = (8, 16, 24, 32, 64, 98, 128, 192 and 256), then =7, for grayscale conditions When the value is 8, the corresponding contrast can be calculated from the grayscale image of the composite surface of the precast concrete component according to equations (2) and (3). and grayscale entropy Correspondingly, when the grayscale level is 16, the corresponding contrast value can also be calculated. and grayscale entropy value Thus, the contrast matrix can be obtained. and grayscale entropy matrix .

[0078] This invention uses equations (2) and (3) above to quickly and accurately calculate the contrast and gray entropy of the image gray co-occurrence matrix under different image gray level conditions, thus providing a basis for the subsequent accurate acquisition of the contrast matrix and gray entropy matrix.

[0079] Further, in step 2), before calculating the image gray-level co-occurrence matrix of the gray-level image of the composite surface of the precast concrete component under different image gray-level conditions, it is also included to set image gray-level conditions. The image gray-level conditions include at least three, with the maximum image gray-level value being 256. For example, the image gray-levels may specifically include 8, 16, 24, 32, 64, 98, 128, 192, and 256. In other embodiments, the number and value of the image gray-level settings only need to be able to represent the contrast and gray-level entropy of the composite surface under different levels of detail information richness, such as the image gray-levels including at least one single-digit, tens-digit, and hundreds-digit image gray-levels.

[0080] Further, in step 3), the weights of contrast and gray-level entropy are calculated using the entropy weight method, including: calculating the contrast of the image gray-level co-occurrence matrix under different image gray-level conditions. and grayscale entropy Normalization is performed. This represents the total number of gray levels in the grayscale image of the composite surface of precast concrete components. The contrast matrix and grayscale entropy matrix are calculated using the following formulas based on the normalized contrast and grayscale entropy:

[0081] (4)

[0082] (5)

[0083] (6)

[0084] (7)

[0085] in, This is the normalized contrast matrix. This is the normalized gray-level entropy matrix. , They are respectively 、 Elements in the matrix In the first Contrast of the gray-level co-occurrence matrix of an image under a given number of gray levels; In the first Gray-level entropy of the gray-level co-occurrence matrix of an image under a given gray level condition;

[0086] The information entropy values ​​of contrast and grayscale entropy are calculated using the following formulas based on the normalized contrast matrix and grayscale entropy matrix:

[0087] (8)

[0088] (9)

[0089] in, The information entropy value of contrast. This represents the information entropy value of the grayscale entropy.

[0090] The present invention uses the above formulas (8) and (9) to quickly and accurately calculate the information entropy values ​​of contrast and grayscale, thereby accurately obtaining the weights of contrast and grayscale based on the information entropy values ​​of contrast and grayscale.

[0091] Preferably, based on the information entropy values ​​of contrast and grayscale entropy calculated under each grayscale level condition, the weights of contrast and grayscale entropy are calculated using the following formulas:

[0092] (10)

[0093] (11)

[0094] in, As the weight of contrast, The weight is the grayscale entropy.

[0095] The weights and grayscale entropy calculated according to the above formulas (10) and (11) can comprehensively represent the contrast and grayscale entropy of the superimposed surface under different levels of detail information, accurately represent the importance of contrast and grayscale entropy under objective conditions, and thus adaptively determine the weights corresponding to contrast and grayscale entropy.

[0096] This invention uses the above formulas (4) to (11), combined with the entropy weight method to quickly and accurately calculate the weight of contrast and the weight of gray entropy. Thus, according to the actual state of contrast and gray entropy in the two-dimensional image of the composite surface of precast concrete components, the weights of contrast and gray entropy are adaptively adjusted. When the information entropy value of contrast is large, the corresponding weight is increased. Correspondingly, when the information entropy value of gray entropy is large, the corresponding weight is increased. This allows for the real-time and accurate determination of the weights of contrast and gray entropy under different working conditions.

[0097] In step 4), the contrast and gray-level entropy of the gray-level image of the composite surface of the precast concrete component are weighted and summed at a specified gray level using the following formula, based on the calculated weights of contrast and gray-level entropy, to calculate the roughness index of the composite surface of the precast concrete component:

[0098] (12)

[0099] in, This refers to the roughness index of the mating surfaces of precast concrete components. As the weight of contrast, The weight of grayscale entropy To achieve the maximum gray level Contrast after normalization under certain conditions Maximum gray level Under the given conditions, the normalized gray-level entropy, for example, if the maximum gray level is 256, then... =256. The higher the gray level, the richer the detail information that can be obtained from image processing. The most accurate roughness index value can be obtained under the condition of maximum gray level.

[0100] This invention uses the above formula (12) to combine the contrast and grayscale entropy states in the actual image and use adaptive weights to... By weighting the contrast and grayscale entropy under specific conditions, the roughness index of the composite surface of precast concrete components is accurately obtained. Compared with the traditional method using fixed weights, this significantly improves the accuracy and reliability of roughness assessment. The roughness index ranges from 0 to 1, with smaller values ​​indicating rougher surfaces. For example, 0-0.3 can be defined as high roughness, 0.3-0.6 as medium roughness, and 0.6-1.0 as low roughness.

[0101] The roughness index calculated based on this invention can further quantify and evaluate the roughness of the composite surface of precast concrete components, and clearly compare the roughness of the composite surface in different embodiments. This invention uses the above method to obtain a roughness index in the range of 0 to 1. This roughness index considers both overall and local roughness, enabling effective quantitative evaluation. Furthermore, because the roughness index has a boundary range, it also allows for quantitative comparison of the roughness of different components, providing a more intuitive and accurate comparison to determine the roughness of different components. For example, if the roughness index of the first set of composite surfaces is calculated to be 0.2 and the roughness index of the second set of composite surfaces is 0.3 using the method of this invention, it can be clearly determined that the roughness of the first set of composite surfaces is greater than that of the second set, and the difference in roughness can be quantified as 0.1, achieving a precise quantitative comparison of the roughness of different components.

[0102] This application also provides a computer device, including a processor and a memory, wherein the memory is used to store a computer program and the processor is used to execute the computer program to perform the method for evaluating the roughness of the composite surface of precast concrete components provided in any of the above embodiments.

[0103] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method for evaluating the roughness of the composite surface of precast concrete components provided in any of the above embodiments.

[0104] In a specific application embodiment of the present invention, the specific process of using the above-described method for evaluating the roughness of the composite surface of precast concrete components based on two-dimensional images is as follows:

[0105] In step 1), a two-dimensional image of the overlapping surface of the precast concrete components is acquired, and the acquired two-dimensional image is processed to obtain a grayscale image of the overlapping surface of the precast concrete components.

[0106] like Figure 2 As shown, two sets of two-dimensional photographs of the overlapping surfaces of precast concrete components were taken. Both sets of precast concrete components are prefabricated concrete eaves panels. High-definition digital cameras were used to capture these images, ensuring that the two-dimensional images of the overlapping surfaces of the precast concrete components have a resolution greater than 20 million pixels. This provides high-resolution two-dimensional images of the overlapping surfaces, guaranteeing the reliability and accuracy of subsequent analysis data.

[0107] Next, image optimization processing is performed based on the acquired two-dimensional photographs of the composite surfaces of the precast concrete components. Preferably, firstly, the background of the two-dimensional image is cropped to remove background impurities from the high-resolution image, leaving only the main body of the composite surfaces of the precast concrete components. Then, a threshold segmentation function (such as the threshold segmentation function included in the OpenCV library of Python) is used to perform grayscale processing on the two-dimensional image after removing the background impurities. Finally, a Gaussian filtering function (such as the Gaussian filtering function included in the OpenCV library) is used to further denoise the two-dimensional image, ultimately obtaining the image shown below. Figure 3 The image shown is a grayscale image of the overlapping surface of the two sets of precast concrete components after optimization.

[0108] In step 2), multiple image gray levels are set. This involves setting the image gray levels to 8, 16, 24, 32, 64, 98, 128, 192, and 256, respectively. Then, based on the different image gray level conditions, a gray-level co-occurrence matrix function (such as the gray-level co-occurrence matrix function in the scikit-image library of Python) is used to read and construct the image gray-level co-occurrence matrix under different image gray level conditions using Equation 1) above.

[0109] Based on the image gray-level co-occurrence matrix under different image gray-level conditions, the contrast of the image gray-level co-occurrence matrix under different image gray-level conditions can be further calculated using Equation 2) above, and the gray-level entropy of the image gray-level co-occurrence matrix under different image gray-level conditions can be calculated using Equation 3) above. After calculation, two sets of contrast and gray-level entropy under different image gray-level conditions can be obtained, as shown in Table 1.

[0110]

[0111] In step 3), the weights of contrast and grayscale entropy are calculated using the entropy weight method, specifically including the following steps;

[0112] In this embodiment, the contrast of the image gray-level co-occurrence matrix shown in Table 1 under different image gray-level conditions is analyzed. and grayscale entropy After normalization, the normalized contrast matrix can be calculated using equations 4) and 5), and the normalized grayscale entropy matrix can be calculated using equations 6) and 7). The matrix elements are shown in Table 2.

[0113]

[0114] Based on the normalized contrast matrix and grayscale entropy matrix The information entropy value of contrast is calculated using Equation 8. The information entropy value of grayscale entropy is calculated using Equation 9. For the first group, the information entropy values ​​for contrast and grayscale entropy are 0.6770 and 0.5690, respectively. For the second group, the information entropy values ​​for contrast and grayscale entropy are 0.4172 and 0.4826, respectively.

[0115] Based on the calculated information entropy values ​​of contrast and grayscale entropy, the weight of contrast can be further calculated using Equation 10. The weights of the grayscale entropy are calculated using Equation 11. For the first group, the weights of contrast and grayscale entropy are 0.4292 and 0.5708, respectively. For the second group, the weights of contrast and grayscale entropy are 0.5364 and 0.4636, respectively.

[0116] In step 4), the weights of the contrast obtained from the above calculations are... ) and the weight of gray entropy ( (and the contrast ratios of the two normalized sets under a grayscale level of 256). and grayscale entropy The roughness index of the two groups can be further calculated using equation (12) to obtain the most accurate roughness index. For the first group, the roughness index is 0.1898, and for the second group, the roughness index is 0.2610. The smaller the value, the rougher the surface.

[0117] This embodiment first evaluates whether the roughness of the images of the first and second sets of overlapping eaves panels meets the requirements given in the "Code for Acceptance of Construction Quality of Concrete Structures" (GB 50204-2015) ("The area of ​​the rough surface should not be less than 80% of the bonding surface, the depth of the rough surface of the precast slab should not be less than 4mm, and the depth of the rough surface of the precast beam end, precast column end, and precast wall end should not be less than 6mm"). If it meets the requirements, the roughness index can be further calculated based on the above method of this invention to further quantify and evaluate its roughness. The experimental results show that the surface roughness of the first set of overlapping eaves panels is greater than that of the second set of overlapping eaves panels, which is consistent with the actual situation. This verifies that this invention can effectively evaluate the roughness of the overlapping surface.

[0118] In summary, the evaluation method of the present invention is simple to operate and has extremely high detection efficiency, and can quickly complete the roughness detection task of a large number of prefabricated component composite surfaces.

[0119] Although the invention has been described with reference to preferred embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, the technical features mentioned in the various embodiments can be combined in any manner, provided there is no structural conflict. The invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for evaluating the roughness of a superimposed surface of a concrete precast member based on a two-dimensional image, characterized by, The method comprises the following steps: Step 1), collecting a two-dimensional image of a composite surface of a concrete prefabricated component, and performing image grayscale processing on the collected two-dimensional image to obtain a grayscale image of the composite surface of the concrete prefabricated component; Step 2), calculating image gray level co-occurrence matrices of the grayscale image of the composite surface of the concrete prefabricated component under different image grayscale conditions; Based on the calculated image gray level co-occurrence matrices, the contrast and the gray level entropy of the image gray level co-occurrence matrices under different image grayscale conditions are calculated; Step 3), according to the contrast and the gray level entropy of the image gray level co-occurrence matrices under different image grayscale conditions, the weights corresponding to the contrast and the gray level entropy are respectively calculated; Step 4), the contrast and the gray level entropy of the grayscale image of the composite surface of the concrete prefabricated component under the maximum grayscale condition are weighted and summed using the calculated weights of the contrast and the gray level entropy, and a roughness index of the composite surface of the concrete prefabricated component is calculated, so as to evaluate the roughness state of the composite surface of the concrete prefabricated component according to the roughness index.

2. The two-dimensional image-based concrete precast member overlay surface roughness evaluation method according to claim 1, characterized by, In step 2), the formula for calculating each element of the image gray level co-occurrence matrices of the grayscale image of the composite surface of the concrete prefabricated component under different image grayscale conditions is: wherein, P (G1, G2) is the probability of the upper gray scale being and the lower gray scale being in the vertical adjacent pixels of the gray scale image of the composite surface of the concrete prefabricated component; N is the total number of pixels of the gray scale image of the composite surface of the concrete prefabricated component; P (G1, G2) is the number of pixel pairs starting from the pixel point with the gray scale of and reaching the pixel point with the gray scale of with the distance of 1 and the direction of 0 degree; N is the number of gray scales of the gray scale image under different image gray scale conditions.

3. The two-dimensional image-based concrete precast member overlay surface roughness evaluation method according to claim 2, characterized by, In step 2), based on the calculated image gray level co-occurrence matrices, the contrast and the gray level entropy of the image gray level co-occurrence matrices under different image grayscale conditions are calculated using the following formula: wherein is the contrast of the image gray level co-occurrence matrix under the first image gray level condition; is the gray level entropy of the image gray level co-occurrence matrix under the first image gray level condition, , denotes the total number of image gray level conditions. The contrast matrix is obtained by calculating the contrast of the lower image gray level co-occurrence matrix The contrast matrix is obtained by calculating the contrast of the lower image gray level co-occurrence matrix The gray entropy matrix is obtained by calculating the gray entropy of the lower image gray level co-occurrence matrix The gray entropy matrix is obtained by calculating the gray entropy of the lower image gray level co-occurrence matrix .

4. The two-dimensional image-based concrete precast member overlay surface roughness evaluation method according to claim 3, characterized by, In step 3), the weights of the contrast and the gray level entropy are calculated using the entropy weight method, which comprises: The image gray-level co-occurrence matrix is ​​used under different image gray-level conditions. Contrast matrix below and grayscale entropy matrix After normalization, the normalized contrast matrix and gray-level entropy matrix are obtained: wherein, is the normalized contrast matrix, is the normalized gray level entropy matrix, , are respectively 、 the elements in the matrix, is the contrast of the image gray level co-occurrence matrix under the condition of the gray level; is the gray level entropy of the image gray level co-occurrence matrix under the condition of the gray level. According to the normalized contrast matrix and the gray level entropy matrix, the information entropy value of the contrast and the information entropy value of the gray level entropy are respectively calculated using the following formula: wherein is the information entropy value of the contrast, is the information entropy value of the gray scale entropy; According to the information entropy value of the contrast and the information entropy value of the gray level entropy, the weight of the contrast and the weight of the gray level entropy are determined.

5. The two-dimensional image-based concrete precast member overlay surface roughness evaluation method according to claim 4, characterized by, According to the information entropy value of the contrast and the information entropy value of the gray level entropy, the weight of the contrast and the weight of the gray level entropy are respectively calculated using the following formula: wherein, is a weight for contrast, is a weight for gray entropy.

6. The two-dimensional image-based concrete precast member overlay surface roughness evaluation method according to any one of claims 1 to 5, characterized by, In step 4), the contrast and the gray level entropy of the grayscale image of the composite surface of the concrete prefabricated component under the maximum grayscale condition are weighted and summed using the calculated weights of the contrast and the gray level entropy using the following formula, and a roughness index of the composite surface of the concrete prefabricated component is calculated: in, It is a roughness index for the overlapping surfaces of precast concrete components. As the weight of contrast, The weights are the grayscale entropy. To achieve the maximum gray level Contrast after normalization under certain conditions To achieve the maximum gray level The gray entropy after normalization under certain conditions.

7. The two-dimensional image-based concrete precast member overlay surface roughness evaluation method according to any one of claims 1 to 5, characterized by, In step 2), before calculating the image gray level co-occurrence matrices of the grayscale image of the composite surface of the concrete prefabricated component under different image grayscale conditions, the image grayscale conditions are set, and the image grayscale conditions comprise at least three or more, and the maximum image grayscale level is 256 levels.

8. The two-dimensional image-based concrete precast member overlay surface roughness evaluation method according to any one of claims 1 to 5, characterized by, In step 1), before performing image grayscale processing on the collected two-dimensional image, the following steps are further included: removing impurity backgrounds in the collected two-dimensional image, retaining the main part of the composite surface of the concrete prefabricated component, and obtaining a two-dimensional image after impurities are removed; performing image grayscale processing on the two-dimensional image after impurities are removed using a threshold segmentation method to form a grayscale image of the composite surface of the concrete prefabricated component; and after performing image grayscale processing, performing image denoising processing on the grayscale image of the composite surface of the concrete prefabricated component using a Gaussian filtering method.

9. A computer device comprising a processor and a memory for storing a computer program, characterized in that, The processor is configured to execute the computer program to perform the method according to any one of claims 1-8.

10. A computer readable storage medium storing a computer program, characterized in that, The computer program, which is executed by a processor, implements the method as claimed in any one of claims 1 to 8.

Images