Building material product quality detection method and system based on image processing

Abstract

The invention discloses a building material product quality detection method and system based on image processing, and relates to the technical field of image data processing, and the method comprises the steps: obtaining building material product information, obtaining building material basic force bearing point information based on the building material actual use scene according to the building material product information, and carrying out the detection of building material quality based on the building material basic force bearing point information; and dividing the building material products to obtain building material classification information. The building material products are divided based on the building material basic stress point information, the building material classification information is obtained, different building material products are accurately divided according to the building material classification information, a data basis is provided for follow-up building material quality detection, the quality detection efficiency is improved, and the quality of the building material products can be accurately detected according to the historical production data of the building material products. The building material quality evaluation model and the crack influence model are obtained, and the building material product quality is accurately evaluated through the building material quality evaluation model and the crack influence model, so that the accuracy and the reliability of a detection result are ensured.

Description

Technical Field

[0001] This invention relates to the field of image data processing technology, specifically to a method and system for quality inspection of building materials products based on image processing. Background Technology

[0002] In recent years, with the continuous acceleration of my country's industrialization and increasingly fierce market competition, upgrading product processing and manufacturing technologies has become an important means for enterprises to enhance their product competitiveness. Especially for building materials, with the increasing complexity and multifunctionality of building designs, the quality requirements for building materials are also becoming higher and higher, as the quality of building materials is directly related to building performance. Therefore, quality inspection of building materials has become an important part of quality management in production.

[0003] Currently, quality inspection of building materials faces several challenges. It's impossible to accurately analyze the process parameters of building materials, or to accurately assess their quality based on data. Often, process parameters are directly substituted into data models for quality evaluation, resulting in ambiguous results that affect the stability and reliability of building material quality and fail to accurately reflect the product's condition. Directly conducting performance tests or collecting defect data for each building material would involve a large data processing volume and low efficiency, hindering timely quality inspection. Summary of the Invention

[0004] To address the aforementioned technical problems, this paper provides a method and system for quality inspection of building materials based on image processing. This technical solution solves the problems mentioned in the background art, such as the inability to accurately analyze the process parameters of building materials and the inability to accurately assess the quality of building materials based on data. Often, process parameters are directly substituted into the data model to assess the quality of building materials, resulting in fuzzy test results that affect the stability and reliability of the quality of building materials and fail to accurately reflect the quality status of building materials. If performance testing or defect data collection is performed directly on each building material, not only is the data processing volume large and the detection efficiency low, but the quality of building materials cannot be detected in a timely manner, thus affecting the timeliness of quality inspection.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for quality inspection of building materials products based on image processing includes: Obtain building material product information, which includes building material product information and building material product specification information; Based on building material product information and actual building material usage scenarios, obtain the basic stress point information of the building materials, whereby the basic stress point of the building materials represents the stress point of the building materials in actual construction. Based on the basic stress point information of building materials, building material products are classified to obtain building material classification information; Obtain historical production data for building materials products; Based on historical production data of building materials, obtain building material quality assessment models and crack impact models; Based on building material product information, obtain building material production process parameter information; Based on the building material production process parameters and building material quality assessment models, the standard maximum withstandable stress is obtained. Acquire image data of building materials and obtain information on cracks in the building materials based on the Canny edge detection algorithm; Based on the standard maximum withstand stress, and according to the building material classification information, crack influence model and building material crack information, the actual maximum withstand stress corresponding to each building material product is obtained. Based on building material quality standards, obtain the maximum stress threshold that building materials can withstand. The quality of building materials is determined based on the actual maximum withstand stress and the maximum withstand stress threshold of the building materials. If the actual maximum withstand stress exceeds the maximum withstand stress threshold of the building materials, the building materials are qualified. If the actual maximum withstand stress does not exceed the maximum withstand stress threshold of the building materials, the building materials are unqualified.

[0006] Preferably, the step of classifying building material products based on the stress point information of the building material foundation to obtain building material classification information specifically includes: Based on the information of the stress points of the building material foundation, and based on the building material design analysis, the proportional coefficient of each stress point is obtained. The proportional coefficient of each stress point represents the ratio of the force corresponding to each stress point of the foundation to the force on the building material product as a whole when force is applied to the building material product. The two foundation stress points with the largest proportionality coefficient are designated as the first stress point and the second stress point. Based on the information of the stress points of the building material foundation, the first stress point, and the second stress point, the secondary action strength coefficient is obtained; The distance between the first stress point and the second stress point is taken as the basic stress transmission distance; The product of the secondary action strength coefficient and the basic stress transmission distance is used as the stress mapping distance; Based on building material product information and stress concentration analysis, stress concentration points are identified. The distance between the stress concentration point and the stress characteristic point of the building material is taken as the stress characteristic distance; Building materials are classified based on stress mapping distance and stress characteristic distance to obtain building material classification information; If the stress mapping distance corresponding to a building material product does not exceed the stress characteristic distance, the building material product is classified as a first-class building material product; if the stress mapping distance corresponding to a building material product exceeds the stress characteristic distance, the building material product is classified as a second-class building material product.

[0007] Preferably, obtaining the secondary force strength coefficient based on the foundation stress point information, the first stress point, and the second stress point specifically includes: The line connecting the first and second points of force application is used as the calibration line of force application. Obtain the direction of the force corresponding to the first and second points of force application; Based on the direction of the force corresponding to the first and second points of application, obtain the equivalent point of application; Based on the force directions corresponding to the first and second force points, and based on the composition of forces, the force characteristic direction is obtained; Based on the equivalent point of application, extend the force along the direction of force characteristics to both ends to obtain the equivalent line of action; The intersection of the equivalent line of action and the calibrated line of force is used to obtain the stress characteristic points of the building material; Based on the stress point information of the building material foundation, the stress points of the building material foundation other than the first and second stress points are regarded as secondary action points; Obtain the distance between each secondary point of action and the stress characteristic point of the building material; The secondary action strength coefficient is obtained based on the distance between the secondary action point and the stress characteristic point of the building material, the action ratio coefficient corresponding to each basic stress point, and the building material product information.

[0008] Preferably, obtaining the equivalent point of application based on the force directions corresponding to the first and second points of application specifically includes: Based on the direction of the force corresponding to the first and second force points, extend the lines of force extension to both ends to obtain the extension lines of the force corresponding to the first and second force points. Based on the extension lines of action corresponding to the first and second points of force application, obtain the equivalent points of action. Where the directions of the forces corresponding to the first and second points of force application are not collinear, the intersection of the extension lines of the forces corresponding to the first and second points of force application is taken as the equivalent point of force application. If the directions of the forces corresponding to the first and second force points are parallel or collinear, then the ratio of the proportional coefficients of the forces corresponding to the first and second force points is used as the position correction coefficient. Using any position on the calibrated force line as the basic point of application, the distance between the basic point of application and the first force point is taken as the first application distance, and the distance between the basic point of application and the second force point is taken as the second application distance; The position of the basic point of action on the calibrated force line is adjusted until the ratio of the first action distance to the second action distance is equal to the position correction coefficient, and the basic point of action is then used as the equivalent point of action.

[0009] Preferably, the step of obtaining the building material quality assessment model and crack influence model based on historical production data of building materials specifically includes: Based on historical production data of building materials, obtain information on historical production process parameters and historical building material product information; Based on building material defect detection, historical building material image information is obtained; Based on historical building material image information, and using the Canny edge detection algorithm, building material defect information is obtained, whereby the building material defect refers to the crack information present in the building material product; Based on building material defect information, historical building material products are divided into benchmark building material products and characteristic building material products to obtain historical building material classification information; If a historical building material product has defects, it is classified as a characteristic building material product; if a historical building material product does not have defects, it is classified as a benchmark building material product. Based on historical building material classification information, the existing neural network model is trained using the historical production process parameters of the benchmark building material product as a benchmark to obtain a building material quality assessment model. The building material quality assessment model is used to assess the maximum stress that the building material can withstand. Based on the building material quality assessment model, a crack impact model is obtained according to the characteristics of the building material products.

[0010] Preferably, obtaining the crack influence model based on the characteristic building material product specifically includes: Based on the building material quality assessment model, the maximum standard stress that the building material can withstand is obtained according to the characteristics of the building material products. Based on the characteristic building material products, obtain the information on the stress points of the building material foundation corresponding to each characteristic building material product; The characteristic building materials products are divided into categories to obtain characteristic building material classification information, which includes a first category of characteristic building materials and a second category of characteristic building materials. Based on the characteristic building material classification information, obtain the stress concentration points and stress characteristic points of each characteristic building material product; The line connecting the stress concentration point and the stress characteristic point of each characteristic building material product is taken as the stress characteristic line of that characteristic building material product. The circular region with the stress characteristic line as its diameter is taken as the stress characteristic region; The center of the stress characteristic region is taken as the stress characteristic center; Based on the building material defect information, the distance between the geometric center of the building material crack and the stress characteristic center is used as the geometric characteristic coefficient; Based on the characteristic building material classification information, and based on the maximum withstand stress test, the actual maximum withstand stress corresponding to each characteristic building material is obtained; Based on the standard maximum withstand stress, the actual maximum withstand stress, building material defect information, and characteristic building material classification information, a crack influence model is obtained; The crack influence model is as follows: In the formula, This represents the actual maximum withstand stress. The standard maximum withstand stress, and The fracture toughness coefficient is... and This is the geometric correction factor. Geometric characteristic coefficients, The length of the crack. This represents the crack depth.

[0011] Furthermore, an image processing-based building material product quality inspection system is proposed to implement the inspection method described above, including: The main control module is used to classify historical building material products into benchmark building material products and characteristic building material products based on building material defect information, obtain historical building material classification information, train an existing neural network model based on the historical building material classification information and the historical production process parameter information corresponding to the benchmark building material products, obtain a building material quality assessment model, obtain a crack influence model based on the building material quality assessment model and characteristic building material products, obtain building material production process parameter information based on building material product information, obtain the standard maximum withstand stress based on the building material production process parameter information and the building material quality assessment model, obtain the standard maximum withstand stress based on the standard maximum withstand stress, obtain the actual maximum withstand stress corresponding to each building material product based on the building material classification information, the crack influence model and building material crack information, obtain the maximum withstand stress threshold of the building material based on the building material quality standard, and determine whether the building material product quality is qualified based on the actual maximum withstand stress and the maximum withstand stress threshold of the building material. The information acquisition module is used to acquire building material product information, including building material product material information and building material product specification information. Based on the building material product information and the actual use scenario of the building material, the module acquires the basic stress point information of the building material, acquires the image data of the building material, acquires the crack information of the building material based on the Canny edge detection algorithm, acquires the historical production data of the building material, acquires the historical production process parameter information and historical building material product information based on the historical production data, acquires the historical building material image information based on the building material defect detection, and acquires the building material defect information based on the historical building material image information and the Canny edge detection algorithm. The evaluation module is used to obtain the action ratio coefficient corresponding to each basic stress point based on the building material foundation stress point information and building material design analysis. The two basic stress points with the largest action ratio coefficients are designated as the first stress point and the second stress point. Based on the building material foundation stress point information, the first stress point, and the second stress point, the secondary action strength coefficient is obtained. The distance between the first stress point and the second stress point is taken as the foundation stress transmission distance. The product of the secondary action strength coefficient and the foundation stress transmission distance is taken as the stress mapping distance. Based on the building material product information and stress concentration analysis, stress concentration points are obtained. The distance between the stress concentration points and the building material stress characteristic points is taken as the stress characteristic distance. Based on the stress mapping distance and the stress characteristic distance, the building material products are classified to obtain building material classification information. The display module interacts with the main control module and is used to output and display information on the stress points of the building material foundation, building material classification information, building material production process parameters, building material crack information, and actual maximum withstandable stress.

[0012] Optionally, the main control module specifically includes: The control unit is used to obtain building material production process parameter information based on building material product information, obtain the standard maximum bearable stress based on the building material production process parameter information and building material quality assessment model, obtain the actual maximum bearable stress corresponding to each building material product based on the standard maximum bearable stress, building material classification information, crack influence model and building material crack information, obtain the building material maximum bearable stress threshold based on building material quality standards, and determine whether the building material product quality is qualified based on the actual maximum bearable stress and the building material maximum bearable stress threshold. An information receiving unit, which interacts with the information acquisition module and the evaluation module, is used to receive data and transmit it to the model training unit. The model training unit is used to classify historical building material products into benchmark building material products and characteristic building material products according to building material defect information, obtain historical building material classification information, train the existing neural network model based on the historical building material classification information and the historical production process parameter information of the benchmark building material products, obtain a building material quality assessment model, and obtain a crack influence model based on the building material quality assessment model and the characteristic building material products.

[0013] Optionally, the information acquisition module specifically includes: The first acquisition unit is used to acquire building material product information, which includes building material product material information and building material product specification information. Based on the building material product information and the actual use scenario of the building material, the first acquisition unit is used to acquire the basic stress point information of the building material, acquire the image data of the building material product, and acquire the crack information of the building material based on the Canny edge detection algorithm. The second acquisition unit is used to acquire historical production data of building materials products. Based on the historical production data of building materials products, it acquires historical production process parameter information and historical building material product information. Based on building material defect detection, it acquires historical building material image information. Based on the historical building material image information, it acquires building material defect information based on the Canny edge detection algorithm.

[0014] Optionally, the evaluation module specifically includes: The first evaluation unit is used to obtain the action ratio coefficient corresponding to each foundation stress point based on the building material foundation stress point information and building material design analysis, and to take the two foundation stress points with the largest action ratio coefficients as the first stress point and the second stress point, and to obtain the secondary action strength coefficient based on the building material foundation stress point information, the first stress point and the second stress point. The second evaluation unit is used to take the distance between the first stress point and the second stress point as the basic stress transmission distance, and the product of the secondary action strength coefficient and the basic stress transmission distance as the stress mapping distance. Based on the building material product information and stress concentration analysis, stress concentration points are obtained, and the distance between the stress concentration points and the stress characteristic points of the building material is taken as the stress characteristic distance. Based on the stress mapping distance and the stress characteristic distance, the building material products are divided to obtain building material classification information.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention proposes a method and system for quality inspection of building materials based on image processing. By classifying building materials based on their fundamental stress points, the method obtains classification information, accurately categorizing different building materials to provide a data foundation for subsequent quality inspection and improves inspection efficiency. Furthermore, by utilizing historical production data, the method acquires a quality assessment model and a crack influence model, enabling accurate quality evaluation of building materials and ensuring the accuracy and reliability of the inspection results. Attached Figure Description

[0016] Figure 1 This is a flowchart of a building material product quality inspection method based on image processing proposed in this invention; Figure 2 This is a flowchart of the building materials classification information acquisition process in this invention; Figure 3 This is a flowchart of the process for obtaining the secondary action intensity coefficient in this invention; Figure 4 This is a flowchart of the crack influence model acquisition process in this invention; Figure 5 This is a structural block diagram of a building material product quality inspection system based on image processing proposed in this invention. Detailed Implementation

[0017] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0018] Reference Figure 1 - Figure 4 As shown in the figure, an image processing-based method for inspecting the quality of building materials in an embodiment of the present invention includes: Obtain building material product information, which includes building material product information and building material product specification information; Based on building material product information and actual building material usage scenarios, obtain the basic stress point information of the building materials, whereby the basic stress point of the building materials represents the stress point of the building materials in actual construction. Based on the basic stress point information of building materials, building material products are classified to obtain building material classification information; Specifically, based on the information of the basic stress points of building materials, building material products are classified to obtain building material classification information, including: Based on the information of the stress points of the building material foundation, and based on the building material design analysis, the proportional coefficient of each stress point is obtained. The proportional coefficient of each stress point represents the ratio of the force corresponding to each stress point of the foundation to the force on the building material product as a whole when force is applied to the building material product. The two foundation stress points with the largest proportionality coefficient are designated as the first stress point and the second stress point. Based on the information of the stress points of the building material foundation, the first stress point, and the second stress point, the secondary action strength coefficient is obtained; The distance between the first stress point and the second stress point is taken as the basic stress transmission distance; The product of the secondary action strength coefficient and the basic stress transmission distance is used as the stress mapping distance; Based on building material product information and stress concentration analysis, stress concentration points are identified. The distance between the stress concentration point and the stress characteristic point of the building material is taken as the stress characteristic distance; Building materials are classified based on stress mapping distance and stress characteristic distance to obtain building material classification information; If the stress mapping distance corresponding to a building material product does not exceed the stress characteristic distance, the building material product is classified as a first-class building material product; if the stress mapping distance corresponding to a building material product exceeds the stress characteristic distance, the building material product is classified as a second-class building material product.

[0019] This scheme extracts the proportional coefficient of the basic stress points to accurately pinpoint key stress points, avoiding interference from secondary factors and improving the specificity of classification. It calculates the stress mapping distance by combining the stress transmission distance and the secondary action intensity coefficient, quantifying stress transmission characteristics; and obtains stress characteristic distances through stress concentration analysis, achieving a concrete comparison between stress concentration and transmission relationships. Based on this, building materials are categorized, ensuring the classification results closely reflect the essence of mechanical properties and reducing subjective bias. The clear distinction between the two categories of building materials provides a differentiated benchmark for subsequent quality testing, facilitating the development of testing strategies for different stress characteristics, improving testing accuracy and efficiency, ensuring that quality assessments align with the actual stress conditions of building materials, and laying a scientific foundation for judging the safety performance of building materials.

[0020] Specifically, based on the stress point information of the building material foundation, the first stress point, and the second stress point, the secondary action strength coefficient is obtained, including: The line connecting the first and second points of force application is used as the calibration line of force application. Obtain the direction of the force corresponding to the first and second points of force application; Based on the direction of the force corresponding to the first and second points of application, obtain the equivalent point of application; Based on the force directions corresponding to the first and second force points, and based on the composition of forces, the force characteristic direction is obtained; Based on the equivalent point of application, extend the force along the direction of force characteristics to both ends to obtain the equivalent line of action; The intersection of the equivalent line of action and the calibrated line of force is used to obtain the stress characteristic points of the building material; Based on the stress point information of the building material foundation, the stress points of the building material foundation other than the first and second stress points are regarded as secondary action points; Obtain the distance between each secondary point of action and the stress characteristic point of the building material; The secondary action strength coefficient is obtained based on the distance between the secondary action point and the stress characteristic point of the building material, the action ratio coefficient corresponding to each basic stress point, and the building material product information.

[0021] This scheme simplifies complex forces into a quantifiable mechanical model by calibrating force lines and equivalent points of application. It extracts characteristic force directions based on the principle of force composition, thus concretizing the mechanical influence of secondary points of application. By introducing distance parameters between secondary points and characteristic points, and correlating the proportionality coefficient with product information to calculate the strength coefficient, it achieves hierarchical quantification of the primary and secondary force relationships, avoiding interference from secondary force points on core testing. This coefficient provides accurate secondary mechanical references for subsequent quality inspection and provides a data foundation for analyzing the influence relationship between secondary forces and stress concentration points in building materials, ensuring the accuracy and reliability of building material classification.

[0022] It should be noted that in this embodiment, for each secondary point of action, the individual contribution value is calculated by combining its force proportion, distance from the feature point, and the building material's elastic modulus. ; In the formula, Indicates the first The coefficient of action of each secondary point of action, Indicates the first The proportionality coefficient of each secondary point of action, The elastic modulus of building materials. As the reference elastic modulus, Indicates the distance of stress transmission in the foundation. Indicates the first The distance between each secondary point of action and the stress characteristic point of the building material; The intensity contribution values ​​of all secondary action points are summed to obtain the secondary action intensity coefficient: ; In the formula, This is the intensity coefficient of the secondary action. The sum of the action ratios of all secondary action points is used for further normalization. This represents the total number of secondary action points; Where there exists a distance between the secondary point of action and the force characteristic point (For concentrated action at close range), the individual contribution value of each secondary action point is adjusted to... ,like (Stress transfer distance of the super-foundation), then the contribution value at that point is corrected to (Reduce interference from distant secondary points); It is understandable that when stress is transmitted outward from the point of application, the attenuation is faster as the distance increases. The influence of secondary points of application on the stress characteristic point decreases with increasing distance. That is, the influence (utility) of the point of application is inversely proportional to its distance from the fulcrum (here, the stress characteristic point). Therefore, each secondary point of application is evaluated by distance and action coefficient, and linear superposition (summation) is performed based on the superposition principle in linear elasticity to obtain the overall influence.

[0023] Specifically, based on the force directions corresponding to the first and second force application points, the equivalent application point is obtained, including: Based on the direction of the force corresponding to the first and second force points, extend the lines of force extension to both ends to obtain the extension lines of the force corresponding to the first and second force points. Based on the extension lines of action corresponding to the first and second points of force application, obtain the equivalent points of action. Where the directions of the forces corresponding to the first and second points of force application are not collinear, the intersection of the extension lines of the forces corresponding to the first and second points of force application is taken as the equivalent point of force application. If the directions of the forces corresponding to the first and second force points are parallel or collinear, then the ratio of the proportional coefficients of the forces corresponding to the first and second force points is used as the position correction coefficient. Using any position on the calibrated force line as the basic point of application, the distance between the basic point of application and the first force point is taken as the first application distance, and the distance between the basic point of application and the second force point is taken as the second application distance; The position of the basic point of action on the calibrated force line is adjusted until the ratio of the first action distance to the second action distance is equal to the position correction coefficient, and the basic point of action is then used as the equivalent point of action.

[0024] In this scheme, for two scenarios—"non-collinear" and "collinear / parallel"—the equivalent action point is determined using the intersection of extended lines and the position correction coefficient adjustment method, respectively. This simplifies the complex multi-point stress on building materials into the concentrated action of a single equivalent point. This simplification preserves the core stress characteristics (such as the direction of force composition when non-collinear) while avoiding redundant calculations in multi-point analysis. For non-collinear stress (such as bidirectional stress at the corner of a building material), the center of force composition is directly locked by the intersection of extended lines, ensuring that the equivalent point reflects the actual superposition effect of the stress. For collinear / parallel stress (such as axial tension of a beam), the position of the equivalent point is dynamically adjusted by the position correction coefficient (the ratio of the action proportion coefficient) to ensure that the distance ratio between the equivalent point and the two stress points matches the force proportion. This adaptive processing covers common stress types of building materials (such as bending, shear, and tension), making the detection method applicable to various building materials such as concrete, steel, and boards. The determination of the equivalent action point provides a clear mechanical reference origin for image processing. During image processing, the stress transmission path (such as the pixel region from the equivalent point to the stress concentration point) can be accurately located based on this point, reducing detection errors caused by blurring of the stress point.

[0025] Obtain historical production data for building materials products; Based on historical production data of building materials, obtain building material quality assessment models and crack impact models; Specifically, based on historical production data of building materials, a building material quality assessment model and a crack impact model are obtained, including: Based on historical production data of building materials, obtain information on historical production process parameters and historical building material product information; Based on building material defect detection, historical building material image information is obtained; Based on historical building material image information, and using the Canny edge detection algorithm, building material defect information is obtained, whereby the building material defect refers to the crack information present in the building material product; Based on building material defect information, historical building material products are divided into benchmark building material products and characteristic building material products to obtain historical building material classification information; If a historical building material product has defects, it is classified as a characteristic building material product; if a historical building material product does not have defects, it is classified as a benchmark building material product. Based on historical building material classification information, the existing neural network model is trained using the historical production process parameters of the benchmark building material product as a benchmark to obtain a building material quality assessment model. The building material quality assessment model is used to assess the maximum stress that the building material can withstand. Based on the building material quality assessment model, a crack impact model is obtained according to the characteristics of the building material products.

[0026] It should be noted that in this solution, the existing neural network model is trained based on the historical production process parameters of the benchmark building material product. Specifically, this includes: From historical building material classification information, extract all sample data labeled as "benchmark building material products" (without cracks or defects), remove characteristic building material products (including cracks) and samples with missing or abnormal data, and match two types of core data for each benchmark building material sample: Input feature data: Historical production process parameters of building materials (such as raw material ratio, molding pressure, curing temperature, curing time, heat treatment parameters, etc., and key parameters should be selected according to the type of building material (concrete / steel / ceramic tile) to avoid redundancy); Output label data: The measured maximum stress that the benchmark building material product can withstand (taken from the performance test records in historical production data, unit: MPa).

[0027] The box plot method was used to remove outliers (such as data exceeding 1.5 times the interquartile range) in process parameters and maximum withstand stress to avoid extreme values ​​interfering with model training. The input process parameters are Z-score standardized to eliminate the dimensional differences between different parameters and improve the model convergence speed. We selected a BP neural network, which is suitable for regression tasks, as the basic model. BP neural networks are good at handling the mapping relationship between high-dimensional process parameters and continuous value output (maximum tolerable stress). They have a simple structure and high training efficiency, and are suitable for the rapid deployment needs of industrial scenarios. Network architecture design: Input layer: Number of neurons = number of key process parameters (e.g., 5 core process parameters, then 5 input neurons). Hidden layers: Set 2-3 hidden layers, with the number of neurons in each layer being 1.5-2 times that of the input layer (e.g., if the input layer has 5 neurons, the hidden layer can have 8 or 6 neurons). The activation function should be ReLU (to avoid gradient vanishing). Output layer: 1 output neuron, activation function is Linear (outputs the maximum stress value that can be continuously withstood). Parameter initialization: The hidden layer weights are initialized using the He normal distribution, and the bias term is initialized to 0; the learning rate (initial value 0.001), the number of iterations (initial value 500 rounds), and the batch size (32 or 64, adjusted according to the sample size) are set.

[0028] The pre-processed benchmark building material samples were divided into the following categories according to a 7:2:1 ratio: Training set (70%): used for model weight updates and feature learning; Validation set (20%): Used to monitor the training process, adjust hyperparameters, and avoid overfitting; Test set (10%): Used for final evaluation of the model's generalization performance, independent of the training process.

[0029] Mean squared error (MSE) is used to quantify the deviation between the model's predicted value and the measured maximum withstandable stress; The Adam optimizer is used to adaptively adjust the learning rate and accelerate model convergence. The training set is input into the model in batches, and the predicted values ​​are calculated through forward propagation and the network weights and biases are updated through back propagation. Every 10 iterations, the loss value is calculated using the validation set. If the loss value on the validation set does not decrease for 20 consecutive iterations, the early stopping mechanism is triggered (training is stopped) to avoid overfitting. If the validation set loss is too high (e.g., RMSE > 5MPa), adjust the learning rate (reduce to 0.0005), the number of hidden layer neurons, or the number of iterations, and retrain until the validation set loss stabilizes and decreases.

[0030] The trained model was independently evaluated using a test set, employing two core metrics: Coefficient of determination : Measure the explanatory power of the model It is considered qualified (the closer to 1, the more accurate the prediction). Root mean square error (RMSE): measures the prediction deviation. RMSE ≤ 3MPa is considered acceptable (adjusted according to building material quality standards, such as RMSE ≤ 2MPa for high-precision building materials). If the evaluation indicators fail to meet the standards, supplement the benchmark building material sample data and retrain.

[0031] Specifically, based on the characteristic building material products, a crack influence model is obtained, including: Based on the building material quality assessment model, the maximum standard stress that the building material can withstand is obtained according to the characteristics of the building material products. Based on the characteristic building material products, obtain the information on the stress points of the building material foundation corresponding to each characteristic building material product; The characteristic building materials products are divided into categories to obtain characteristic building material classification information, which includes a first category of characteristic building materials and a second category of characteristic building materials. Based on the characteristic building material classification information, obtain the stress concentration points and stress characteristic points of each characteristic building material product; The line connecting the stress concentration point and the stress characteristic point of each characteristic building material product is taken as the stress characteristic line of that characteristic building material product. The circular region with the stress characteristic line as its diameter is taken as the stress characteristic region; The center of the stress characteristic region is taken as the stress characteristic center; Based on the building material defect information, the distance between the geometric center of the building material crack and the stress characteristic center is used as the geometric characteristic coefficient; Based on the characteristic building material classification information, and based on the maximum withstand stress test, the actual maximum withstand stress corresponding to each characteristic building material is obtained; Based on the standard maximum withstand stress, the actual maximum withstand stress, building material defect information, and characteristic building material classification information, a crack influence model is obtained; The crack influence model is as follows: In the formula, This represents the actual maximum withstand stress. The standard maximum withstand stress, and The fracture toughness coefficient is... and This is the geometric correction factor. Geometric characteristic coefficients, The length of the crack. This represents the crack depth.

[0032] This solution achieves a fully intelligent upgrade of the entire chain from defect identification to performance evaluation in building material quality inspection through data-driven modeling, defect quantification analysis, and multi-dimensional feature fusion. It accurately extracts crack defects using the Canny edge detection algorithm, categorizes building materials into benchmark / feature products based on historical production data, constructs a training set with clearly defined positive and negative samples, trains the neural network using the process parameters of defect-free products, and establishes a quality assessment model under "ideal conditions." This provides a benchmark for defect impact analysis. For feature products containing cracks, the solution further subdivides them into categories (Category 1 / Category 2) according to stress characteristics, avoiding model bias caused by mixed defect types and ensuring the accuracy and reliability of quality inspection.

[0033] It is understandable that the impact of cracks on the quality needs to be considered when testing the quality of building materials. However, the degree of impact of different crack data on the quality of building materials varies greatly. If the quality of building materials is analyzed by comprehensively considering multiple crack data, the data volume is large and the testing efficiency is low. Therefore, this solution classifies building materials and selects different crack data for different building materials. For the first type of building materials (stress mapping distance ≤ stress characteristic distance), the stress concentration point is close to the stress characteristic point, the stress transmission path is short and concentrated, and the crack propagation direction is more likely to be along the material surface (such as transverse cracks). In this case, the crack length is the key parameter controlling fracture. For the second type of building materials (stress mapping distance > stress characteristic distance), the stress concentration point is far from the stress characteristic point, the stress transmission path is dispersed but may penetrate deep into the material, and the crack is more likely to propagate along the thickness direction (such as longitudinal cracks). In this case, the crack depth has a greater impact on fracture. Therefore, this solution classifies and judges based on the Irwin fracture toughness formula, that is: ; In the formula, for Open-type fracture toughness (an inherent material property, in units of...) This indicates the material's ability to resist crack propagation. This is a geometric correction factor (dimensionless), which is related to the material shape, crack location, and loading method, and corrects for stress concentration effects around the crack. The nominal stress on the material. This is a characteristic parameter of the crack (unit: m), usually the crack half-length (through crack) or crack length (surface crack). By transforming the Irwin fracture toughness formula, we obtain: ; In the formula, Indicates fracture toughness The material critical value; Among them, the inherent fracture toughness of the material Maximum stress that the material can withstand Binding, that is: ; In summary, a crack influence model was obtained.

[0034] It should be noted that in this embodiment, the classification steps for the characteristic building material products are the same as those for the classification of building material products to obtain building material classification information. The initial values ​​are referenced from the standard values ​​in "Test Method for Fracture Toughness of Metallic Materials" (GB / T4161) or "Code for Design of Concrete Structures" (GB50010). The fracture toughness coefficient is obtained by fitting the standard maximum withstand stress, actual maximum withstand stress, building material defect information, and characteristic building material classification information into the crack influence model. and Geometric correction factor and .

[0035] Based on building material product information, obtain building material production process parameter information; Based on the building material production process parameters and building material quality assessment models, the standard maximum withstandable stress is obtained. Acquire image data of building materials and obtain information on cracks in the building materials based on the Canny edge detection algorithm; Based on the standard maximum withstand stress, and according to the building material classification information, crack influence model and building material crack information, the actual maximum withstand stress corresponding to each building material product is obtained. It should be noted that in this embodiment, the first category of building material products in the building material classification information is compared with the crack influence model. Correspondingly, the second category of building materials in the building materials classification information is matched with the crack influence model. Correspondingly, the steps for obtaining the geometric characteristic coefficients of building materials products are the same as those for obtaining the geometric characteristic coefficients of characteristic building materials products; Based on building material quality standards, obtain the maximum stress threshold that building materials can withstand. The quality of building materials is determined based on the actual maximum withstand stress and the maximum withstand stress threshold of the building materials. If the actual maximum withstand stress exceeds the maximum withstand stress threshold of the building materials, the building materials are qualified. If the actual maximum withstand stress does not exceed the maximum withstand stress threshold of the building materials, the building materials are unqualified.

[0036] Reference Figure 5 As shown, further, combining the above-mentioned image processing-based building material product quality inspection method, an image processing-based building material product quality inspection system is proposed, including: The main control module is used to classify historical building material products into benchmark building material products and characteristic building material products based on building material defect information, obtain historical building material classification information, train an existing neural network model based on the historical building material classification information and the historical production process parameter information corresponding to the benchmark building material products, obtain a building material quality assessment model, obtain a crack influence model based on the building material quality assessment model and characteristic building material products, obtain building material production process parameter information based on building material product information, obtain the standard maximum withstand stress based on the building material production process parameter information and the building material quality assessment model, obtain the standard maximum withstand stress based on the standard maximum withstand stress, obtain the actual maximum withstand stress corresponding to each building material product based on the building material classification information, the crack influence model and building material crack information, obtain the maximum withstand stress threshold of the building material based on the building material quality standard, and determine whether the building material product quality is qualified based on the actual maximum withstand stress and the maximum withstand stress threshold of the building material. The information acquisition module is used to acquire building material product information, including building material product material information and building material product specification information. Based on the building material product information and the actual use scenario of the building material, the module acquires the basic stress point information of the building material, acquires the image data of the building material, acquires the crack information of the building material based on the Canny edge detection algorithm, acquires the historical production data of the building material, acquires the historical production process parameter information and historical building material product information based on the historical production data, acquires the historical building material image information based on the building material defect detection, and acquires the building material defect information based on the historical building material image information and the Canny edge detection algorithm. The evaluation module is used to obtain the action ratio coefficient corresponding to each basic stress point based on the building material foundation stress point information and building material design analysis. The two basic stress points with the largest action ratio coefficients are designated as the first stress point and the second stress point. Based on the building material foundation stress point information, the first stress point, and the second stress point, the secondary action strength coefficient is obtained. The distance between the first stress point and the second stress point is taken as the foundation stress transmission distance. The product of the secondary action strength coefficient and the foundation stress transmission distance is taken as the stress mapping distance. Based on the building material product information and stress concentration analysis, stress concentration points are obtained. The distance between the stress concentration points and the building material stress characteristic points is taken as the stress characteristic distance. Based on the stress mapping distance and the stress characteristic distance, the building material products are classified to obtain building material classification information. The display module interacts with the main control module and is used to output and display information on the stress points of the building material foundation, building material classification information, building material production process parameters, building material crack information, and actual maximum withstandable stress.

[0037] The main control module specifically includes: The control unit is used to obtain building material production process parameter information based on building material product information, obtain the standard maximum bearable stress based on the building material production process parameter information and building material quality assessment model, obtain the actual maximum bearable stress corresponding to each building material product based on the standard maximum bearable stress, building material classification information, crack influence model and building material crack information, obtain the building material maximum bearable stress threshold based on building material quality standards, and determine whether the building material product quality is qualified based on the actual maximum bearable stress and the building material maximum bearable stress threshold. An information receiving unit, which interacts with the information acquisition module and the evaluation module, is used to receive data and transmit it to the model training unit. The model training unit is used to classify historical building material products into benchmark building material products and characteristic building material products according to building material defect information, obtain historical building material classification information, train the existing neural network model based on the historical building material classification information and the historical production process parameter information of the benchmark building material products, obtain a building material quality assessment model, and obtain a crack influence model based on the building material quality assessment model and the characteristic building material products.

[0038] The information acquisition module specifically includes: The first acquisition unit is used to acquire building material product information, which includes building material product material information and building material product specification information. Based on the building material product information and the actual use scenario of the building material, the first acquisition unit is used to acquire the basic stress point information of the building material, acquire the image data of the building material product, and acquire the crack information of the building material based on the Canny edge detection algorithm. The second acquisition unit is used to acquire historical production data of building materials products. Based on the historical production data of building materials products, it acquires historical production process parameter information and historical building material product information. Based on building material defect detection, it acquires historical building material image information. Based on the historical building material image information, it acquires building material defect information based on the Canny edge detection algorithm.

[0039] The evaluation module specifically includes: The first evaluation unit is used to obtain the action ratio coefficient corresponding to each foundation stress point based on the building material foundation stress point information and building material design analysis, and to take the two foundation stress points with the largest action ratio coefficients as the first stress point and the second stress point, and to obtain the secondary action strength coefficient based on the building material foundation stress point information, the first stress point and the second stress point. The second evaluation unit is used to take the distance between the first stress point and the second stress point as the basic stress transmission distance, and the product of the secondary action strength coefficient and the basic stress transmission distance as the stress mapping distance. Based on the building material product information and stress concentration analysis, stress concentration points are obtained, and the distance between the stress concentration points and the stress characteristic points of the building material is taken as the stress characteristic distance. Based on the stress mapping distance and the stress characteristic distance, the building material products are divided to obtain building material classification information.

[0040] In summary, the advantages of this invention are as follows: by classifying building material products based on the stress point information of the building material foundation, building material classification information is obtained. This classification information allows for accurate segmentation of different building material products, providing a data foundation for subsequent building material quality testing and improving testing efficiency. Furthermore, by utilizing historical production data of building material products, a building material quality assessment model and a crack influence model are obtained. These models are then used to accurately assess the quality of building material products, ensuring the accuracy and reliability of the test results.

[0041] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A method for quality inspection of building materials products based on image processing, characterized in that, include: Obtain building material product information, which includes building material product information and building material product specification information; Based on building material product information and actual building material usage scenarios, obtain the basic stress point information of the building materials. The basic stress point of the building materials represents the stress point of the building materials in actual construction. Based on the basic stress point information of building materials, building material products are classified to obtain building material classification information; Obtain historical production data for building materials products; Based on historical production data of building materials, obtain building material quality assessment models and crack impact models; Based on building material product information, obtain building material production process parameter information; Based on the building material production process parameters and building material quality assessment models, the standard maximum withstandable stress is obtained. Acquire image data of building materials and obtain information on cracks in the building materials based on the Canny edge detection algorithm; Based on the standard maximum withstand stress, and according to the building material classification information, crack influence model and building material crack information, the actual maximum withstand stress corresponding to each building material product is obtained. Based on building material quality standards, obtain the maximum stress threshold that building materials can withstand. The quality of building materials is determined based on the actual maximum withstand stress and the maximum withstand stress threshold of the building materials. If the actual maximum withstand stress exceeds the maximum withstand stress threshold of the building materials, the building materials are qualified. If the actual maximum withstand stress does not exceed the maximum withstand stress threshold of the building materials, the building materials are unqualified.

2. The method for quality inspection of building materials products based on image processing according to claim 1, characterized in that, The process of classifying building material products based on the fundamental stress point information to obtain building material classification information specifically includes: Based on the information of the stress points of the building material foundation, and based on the building material design analysis, the proportional coefficient of each stress point is obtained. The proportional coefficient of each stress point represents the ratio of the force corresponding to each stress point of the foundation to the force on the building material product as a whole when force is applied to the building material product. The two foundation stress points with the largest proportionality coefficient are designated as the first stress point and the second stress point. Based on the information of the stress points of the building material foundation, the first stress point, and the second stress point, the secondary action strength coefficient is obtained; The distance between the first stress point and the second stress point is taken as the basic stress transmission distance; The product of the secondary action strength coefficient and the basic stress transmission distance is used as the stress mapping distance; Based on building material product information and stress concentration analysis, stress concentration points are identified. The distance between the stress concentration point and the stress characteristic point of the building material is taken as the stress characteristic distance; Building materials are classified based on stress mapping distance and stress characteristic distance to obtain building material classification information; If the stress mapping distance corresponding to a building material product does not exceed the stress characteristic distance, the building material product is classified as a first-class building material product; if the stress mapping distance corresponding to a building material product exceeds the stress characteristic distance, the building material product is classified as a second-class building material product.

3. The method for quality inspection of building materials products based on image processing according to claim 2, characterized in that, The process of obtaining the secondary action strength coefficient based on the stress point information of the building material foundation, the first stress point, and the second stress point specifically includes: The line connecting the first and second points of force application is used as the calibration line of force application. Obtain the direction of the force corresponding to the first and second points of force application; Based on the direction of the force corresponding to the first and second points of application, obtain the equivalent point of application; Based on the force directions corresponding to the first and second force points, and based on the composition of forces, the force characteristic direction is obtained; Based on the equivalent point of application, extend the force along the direction of force characteristics to both ends to obtain the equivalent line of action; The intersection of the equivalent line of action and the calibrated line of force is used to obtain the stress characteristic points of the building material; Based on the stress point information of the building material foundation, the stress points of the building material foundation other than the first and second stress points are regarded as secondary action points; Obtain the distance between each secondary point of action and the stress characteristic point of the building material; The secondary action strength coefficient is obtained based on the distance between the secondary action point and the stress characteristic point of the building material, the action ratio coefficient corresponding to each basic stress point, and the building material product information.

4. The image processing-based method for quality inspection of building materials products according to claim 3, characterized in that, The step of obtaining the equivalent point of application based on the force directions corresponding to the first and second points of application specifically includes: Based on the direction of the force corresponding to the first and second force points, extend the lines of force extension to both ends to obtain the extension lines of the force corresponding to the first and second force points. Based on the extension lines of action corresponding to the first and second points of force application, obtain the equivalent points of action. Where the directions of the forces corresponding to the first and second points of force application are not collinear, the intersection of the extension lines of the forces corresponding to the first and second points of force application is taken as the equivalent point of force application. If the directions of the forces corresponding to the first and second force points are parallel or collinear, then the ratio of the proportional coefficients of the forces corresponding to the first and second force points is used as the position correction coefficient. Using any position on the calibrated force line as the basic point of application, the distance between the basic point of application and the first force point is taken as the first application distance, and the distance between the basic point of application and the second force point is taken as the second application distance; The position of the basic point of action on the calibrated force line is adjusted until the ratio of the first action distance to the second action distance is equal to the position correction coefficient, and the basic point of action is then used as the equivalent point of action.

5. The method for quality inspection of building materials products based on image processing according to claim 4, characterized in that, The process of obtaining a building material quality assessment model and a crack impact model based on historical production data of building materials specifically includes: Based on historical production data of building materials, obtain information on historical production process parameters and historical building material product information; Based on building material defect detection, historical building material image information is obtained; Based on historical building material image information, and using the Canny edge detection algorithm, building material defect information is obtained, whereby the building material defect refers to the crack information present in the building material product; Based on building material defect information, historical building material products are divided into benchmark building material products and characteristic building material products to obtain historical building material classification information; If a historical building material product has defects, it is classified as a characteristic building material product; if a historical building material product does not have defects, it is classified as a benchmark building material product. Based on historical building material classification information, the existing neural network model is trained using the historical production process parameters of the benchmark building material product as a benchmark to obtain a building material quality assessment model. The building material quality assessment model is used to assess the maximum stress that the building material can withstand. Based on the building material quality assessment model, a crack impact model is obtained according to the characteristics of the building material products.

6. The image processing-based method for quality inspection of building materials products according to claim 5, characterized in that, The step of obtaining the crack influence model based on the characteristic building material product specifically includes: Based on the building material quality assessment model, the maximum standard stress that the building material can withstand is obtained according to the characteristics of the building material products. Based on the characteristic building material products, obtain the information on the stress points of the building material foundation corresponding to each characteristic building material product; The characteristic building materials products are divided into categories to obtain characteristic building material classification information, which includes a first category of characteristic building materials and a second category of characteristic building materials. Based on the characteristic building material classification information, obtain the stress concentration points and stress characteristic points of each characteristic building material product; The line connecting the stress concentration point and the stress characteristic point of each characteristic building material product is taken as the stress characteristic line of that characteristic building material product. The circular region with the stress characteristic line as its diameter is taken as the stress characteristic region; The center of the stress characteristic region is taken as the stress characteristic center; Based on the building material defect information, the distance between the geometric center of the building material crack and the stress characteristic center is used as the geometric characteristic coefficient; Based on the characteristic building material classification information, and based on the maximum withstand stress test, the actual maximum withstand stress corresponding to each characteristic building material is obtained; Based on the standard maximum withstand stress, the actual maximum withstand stress, building material defect information, and characteristic building material classification information, a crack influence model is obtained; The crack influence model is as follows: In the formula, This represents the actual maximum withstand stress. The standard maximum withstand stress, and The fracture toughness coefficient is... and This is the geometric correction factor. Geometric characteristic coefficients The length of the crack. This represents the crack depth.

7. A building materials product quality inspection system based on image processing, used to implement the inspection method as described in any one of claims 1-6, characterized in that, include: The main control module is used to classify historical building material products into benchmark building material products and characteristic building material products based on building material defect information, obtain historical building material classification information, train an existing neural network model based on the historical building material classification information and the historical production process parameter information corresponding to the benchmark building material products, obtain a building material quality assessment model, obtain a crack influence model based on the building material quality assessment model and characteristic building material products, obtain building material production process parameter information based on building material product information, obtain the standard maximum withstand stress based on the building material production process parameter information and the building material quality assessment model, obtain the standard maximum withstand stress based on the standard maximum withstand stress, obtain the actual maximum withstand stress corresponding to each building material product based on the building material classification information, the crack influence model and building material crack information, obtain the maximum withstand stress threshold of the building material based on the building material quality standard, and determine whether the building material product quality is qualified based on the actual maximum withstand stress and the maximum withstand stress threshold of the building material. The information acquisition module is used to acquire building material product information, including building material product material information and building material product specification information. Based on the building material product information and the actual use scenario of the building material, the module acquires the basic stress point information of the building material, acquires the image data of the building material, acquires the crack information of the building material based on the Canny edge detection algorithm, acquires the historical production data of the building material, acquires the historical production process parameter information and historical building material product information based on the historical production data, acquires the historical building material image information based on the building material defect detection, and acquires the building material defect information based on the historical building material image information and the Canny edge detection algorithm. The evaluation module is used to obtain the action ratio coefficient corresponding to each basic stress point based on the building material foundation stress point information and building material design analysis. The two basic stress points with the largest action ratio coefficients are designated as the first stress point and the second stress point. Based on the building material foundation stress point information, the first stress point, and the second stress point, the secondary action strength coefficient is obtained. The distance between the first stress point and the second stress point is taken as the foundation stress transmission distance. The product of the secondary action strength coefficient and the foundation stress transmission distance is taken as the stress mapping distance. Based on the building material product information and stress concentration analysis, stress concentration points are obtained. The distance between the stress concentration points and the building material stress characteristic points is taken as the stress characteristic distance. Based on the stress mapping distance and the stress characteristic distance, the building material products are classified to obtain building material classification information. The display module interacts with the main control module and is used to output and display information on the stress points of the building material foundation, building material classification information, building material production process parameters, building material crack information, and the actual maximum bearable stress.

8. The image processing-based building materials product quality inspection system according to claim 7, characterized in that, The main control module specifically includes: The control unit is used to obtain building material production process parameter information based on building material product information, obtain the standard maximum bearable stress based on the building material production process parameter information and building material quality assessment model, obtain the actual maximum bearable stress corresponding to each building material product based on the standard maximum bearable stress, building material classification information, crack influence model and building material crack information, obtain the building material maximum bearable stress threshold based on building material quality standards, and determine whether the building material product quality is qualified based on the actual maximum bearable stress and the building material maximum bearable stress threshold. An information receiving unit, which interacts with the information acquisition module and the evaluation module, is used to receive data and transmit it to the model training unit. The model training unit is used to classify historical building material products into benchmark building material products and characteristic building material products according to building material defect information, obtain historical building material classification information, train the existing neural network model based on the historical building material classification information and the historical production process parameter information of the benchmark building material products, obtain a building material quality assessment model, and obtain a crack influence model based on the building material quality assessment model and the characteristic building material products.

9. A building materials product quality inspection system based on image processing according to claim 7, characterized in that, The information acquisition module specifically includes: The first acquisition unit is used to acquire building material product information, which includes building material product material information and building material product specification information. Based on the building material product information and the actual use scenario of the building material, the first acquisition unit is used to acquire the basic stress point information of the building material, acquire the image data of the building material product, and acquire the crack information of the building material based on the Canny edge detection algorithm. The second acquisition unit is used to acquire historical production data of building materials products. Based on the historical production data of building materials products, it acquires historical production process parameter information and historical building material product information. Based on building material defect detection, it acquires historical building material image information. Based on the historical building material image information, it acquires building material defect information based on the Canny edge detection algorithm.

10. A building materials product quality inspection system based on image processing according to claim 7, characterized in that, The evaluation module specifically includes: The first evaluation unit is used to obtain the action ratio coefficient corresponding to each foundation stress point based on the building material foundation stress point information and building material design analysis, and to take the two foundation stress points with the largest action ratio coefficients as the first stress point and the second stress point, and to obtain the secondary action strength coefficient based on the building material foundation stress point information, the first stress point and the second stress point. The second evaluation unit is used to take the distance between the first stress point and the second stress point as the basic stress transmission distance, and the product of the secondary action strength coefficient and the basic stress transmission distance as the stress mapping distance. Based on the building material product information and stress concentration analysis, stress concentration points are obtained, and the distance between the stress concentration points and the stress characteristic points of the building material is taken as the stress characteristic distance. Based on the stress mapping distance and the stress characteristic distance, the building material products are divided to obtain building material classification information.

Images