Intelligent Detection Method and System for Surface Defects of Building Materials

Abstract

The present invention discloses an intelligent detection method and system for surface defects of building materials, relating to the technical field of data processing. The method includes: obtaining K batches and the quantities of K batches of target building materials; obtaining K initial sampling sample sets, and performing surface defect detection to obtain K first-batch reliability factors; performing volume analysis to determine K second-batch reliability factors; obtaining K conditional sampling sample sets; performing surface defect detection to obtain a set of surface defect detection results of K conditional sampling samples; performing homogeneous aggregation to obtain K clustering detection result clusters; determining a set of median values of K clustering defect feature sets; performing cross-batch weighted calculation according to a preset weight to obtain a set of weighted median values of clustering defect features, and using the set as the surface defect detection results of K. The technical problem of low detection efficiency and accuracy of surface defects of building materials in the prior art is solved, and the technical effect of improving detection efficiency and accuracy is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to an intelligent detection method and system for surface defects of building materials. Background Art

[0002] In the production and application process of building materials, surface defects are important factors affecting product quality and performance. For example, surface defects such as cracks, pores, and scratches may lead to a decrease in the structural strength and protective performance of the materials, thereby affecting their service life and safety. At present, traditional surface defect detection methods mainly rely on manual detection or automated detection means based on a single algorithm. On the one hand, manual detection is inefficient and highly subjective due to the experience and judgment of the inspectors, making it difficult to ensure the objectivity and consistency of the detection results. On the other hand, automated detection technologies based on a single algorithm are insufficient in adaptability when facing different batches or diverse types of building materials, and it is difficult to balance detection accuracy and generalization ability. Summary of the Invention

[0003] This application provides an intelligent detection method and system for surface defects of building materials, which solves the technical problem of low detection efficiency and accuracy of surface defects of building materials in the prior art.

[0004] In view of the above problems, this application provides an intelligent detection method and system for surface defects of building materials.

[0005] In the first aspect of this application, an intelligent detection method for surface defects of building materials is provided. The method includes:

[0006] Obtain K batches and the quantities of K batches of target building materials, where K is an integer greater than or equal to 1; respectively perform initial sample extraction of a preset extraction quantity on the K batches in a random sampling manner to obtain K initial extraction sample sets, and perform surface defect detection on the K initial extraction sample sets, and obtain K first-batch reliability factors according to the detection results; perform volume analysis based on the quantities of the K batches to determine K second-batch reliability factors; perform conditional sampling on the K batches according to the K first-batch reliability factors and the K second-batch reliability factors to obtain K conditional sampling sample sets; traverse the K conditional sampling sample sets to perform surface defect detection to obtain a set of surface defect detection results of the K conditional sampling samples; perform homogeneous aggregation on the set of surface defect detection results of the K conditional sampling samples to obtain K clustering detection result clusters, each clustering detection result cluster includes multiple sets of surface defect detection results of clustering conditional sampling samples, and each set of surface defect detection results of clustering conditional sampling samples includes multiple surface defect detection results of conditional sampling samples belonging to the same surface defect type; perform defect feature value identification on the K clustering detection result clusters, and perform centralized analysis of the identification results to determine K sets of centralized values of clustering defect features; perform cross-batch weighted calculation on the K sets of centralized values of clustering defect features according to a preset weight to obtain a weighted set of centralized values of clustering defect features, and use the weighted set of centralized values of clustering defect features as the surface defect detection results of the K batches.

[0007] In a second aspect of the present application, there is provided an intelligent surface defect detection system for building materials, and the system includes:

[0008] A data acquisition module for acquiring K batches and the quantities of K batches of target building materials, where K is an integer greater than or equal to 1; a first sampling module for respectively extracting initial samples of a preset extraction quantity from the K batches by means of random sampling to obtain K initial extraction sample sets, and performing surface defect detection on the K initial extraction sample sets to obtain K first batch reliability factors based on the detection results; an analysis module for performing volume analysis based on the quantities of the K batches to determine K second batch reliability factors; a second sampling module for performing conditional sampling on the K batches according to the K first batch reliability factors and the K second batch reliability factors to obtain K conditional sampling sample sets; a defect detection module for traversing the K conditional sampling sample sets to perform surface defect detection to obtain a set of surface defect detection results of the K conditional sampling samples; an aggregation module for performing homogeneous aggregation on the set of surface defect detection results of the K conditional sampling samples to obtain K clustering detection result clusters, each clustering detection result cluster including multiple clustering conditional sampling sample surface defect detection result sets, and each clustering conditional sampling sample surface defect detection result set including multiple conditional sampling sample surface defect detection results belonging to the same surface defect type; an identification module for identifying defect characteristic values of the K clustering detection result clusters and performing centralized analysis on the identification results to determine K clustering defect characteristic centralized value sets; a calculation module for performing cross-batch weighted calculation on the K clustering defect characteristic centralized value sets according to a preset weight to obtain a weighted clustering defect characteristic centralized value set, and using the weighted clustering defect characteristic centralized value set as the surface defect detection results of the K batches.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] First, obtain K batches and the quantities of K batches of target building materials, where K is an integer greater than or equal to 1. Then, respectively perform initial sample extraction of a preset extraction quantity on the K batches in a random sampling manner to obtain K initial extraction sample sets, and perform surface defect detection on the K initial extraction sample sets to obtain K first-batch reliability factors based on the detection results. At the same time, conduct volume analysis based on the quantities of the K batches to determine K second-batch reliability factors. Further, perform conditional sampling on the K batches according to the K first-batch reliability factors and the K second-batch reliability factors to obtain K conditional sampling sample sets; and traverse the K conditional sampling sample sets to perform surface defect detection to obtain a set of surface defect detection results for the K conditional sampling samples. Then, perform homogeneous aggregation on the set of surface defect detection results for the K conditional sampling samples to obtain K clustering detection result clusters, each clustering detection result cluster includes multiple sets of surface defect detection results for clustering conditional sampling samples, and each set of surface defect detection results for clustering conditional sampling samples includes multiple surface defect detection results of conditional sampling samples belonging to the same surface defect type. Next, identify defect characteristic values for the K clustering detection result clusters, and conduct centralized analysis of the identification results to determine K sets of centralized values of clustering defect characteristics. Finally, perform cross-batch weighted calculation on the K sets of centralized values of clustering defect characteristics according to a preset weight to obtain a weighted set of centralized values of clustering defect characteristics, and use the weighted set of centralized values of clustering defect characteristics as the surface defect detection results for the K batches. This solves the technical problem of low efficiency and accuracy in surface defect detection of building materials in the prior art, and achieves the technical effect of improving detection efficiency and accuracy. Description of the Drawings

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following-described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0012] Figure 1 Schematic flowchart of the intelligent detection method for surface defects of building materials provided by the embodiment of the present application.

[0013] Figure 2 Schematic structural diagram of the intelligent detection system for surface defects of building materials provided by the embodiment of the present application.

[0014] Description of the reference numerals: data acquisition module 11, first sampling module 12, analysis module 13, second sampling module 14, defect detection module 15, aggregation module 16, identification module 17, calculation module 18. Detailed Embodiments

[0015] By providing an intelligent detection method and system for surface defects of building materials, the present application solves the technical problems of low detection efficiency and accuracy of surface defects of building materials in the prior art.

[0016] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0017] It should be noted that the terms "include" and "have" are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server including a series of steps or units is not necessarily limited to those clearly listed steps or units, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices.

[0018] Embodiment 1, as Figure 1 shown, the present application provides an intelligent detection method for surface defects of building materials, wherein the method includes:

[0019] Obtain K batches and the quantity of K batches of target building materials, where K is an integer greater than or equal to 1.

[0020] When performing intelligent detection of surface defects of building materials, it is necessary to determine K batches and the quantity of K batches of target building materials to be detected, where K is an integer greater than or equal to 1.

[0021] Respectively perform initial sample extraction of a preset extraction quantity on the K batches in a random sampling manner to obtain K initial extraction sample sets, and perform surface defect detection on the K initial extraction sample sets, and obtain K first batch reliability factors according to the detection results.

[0022] For each batch (a total of K batches), perform random sampling according to the preset extraction quantity, that is, randomly select a certain number of building material samples from each batch to form an initial extraction sample set. Through the random sampling process, K initial extraction sample sets are obtained. For the initial sample set of each batch, use the surface defect detection device to perform detection, and record the detection results (such as defect type, defect severity, etc.) of each sample. According to the results of the surface defect detection, perform quality evaluation on the initial extraction sample set of each batch. The proportion of defective samples, the severity of defects and other indicators in this batch can be calculated, and then a first batch reliability factor is obtained based on these indicators. The first batch reliability factor reflects the reliability degree of the building materials in this batch in terms of surface quality.

[0023] Furthermore, obtain K sets of initial extraction samples, perform surface defect detection on the K sets of initial extraction samples, and obtain K first-batch reliability factors according to the detection results, including:

[0024] Perform surface defect detection on the K sets of initial extraction samples to obtain K sets of initial extraction sample detection results; traverse the K sets of initial extraction sample detection results for result stability analysis to obtain K stability factors; respectively take the ratios of the K stability factors to the sum of the K stability factors as the K first-batch reliability factors.

[0025] Preferably, perform surface defect detection on each sample in the K sets of initial extraction samples, and obtain K sets of initial extraction sample detection results according to the detection results; traverse the K sets of initial extraction sample detection results, perform stability analysis on the detection results in each set, and the variance of all sample detection results within each set of initial extraction sample detection results can be calculated, and the reciprocal of the variance is used as the stability factor. The larger the variance, the greater the difference in the detection results of the samples within the batch, and the lower the stability. On the contrary, the smaller the variance, the higher the stability. In order to quantify the reliability of each batch, divide the stability factor of each batch by the sum of the stability factors of all batches, thereby obtaining K first-batch reliability factors.

[0026] Based on the K batch quantities, perform volume analysis to determine K second-batch reliability factors.

[0027] Respectively divide the K batch quantities by the sum of the K batch quantities to obtain K second-batch reliability factors. The second-batch reliability factors reflect the importance of each batch. The larger the batch quantity, the higher the second-batch reliability factor.

[0028] According to the K first-batch reliability factors and the K second-batch reliability factors, perform conditional sampling on the K batches to obtain K sets of conditional sampling samples.

[0029] According to the needs of actual applications, weights are assigned to the reliability factors of the first batch and the reliability factors of the second batch. These weights reflect the importance of the two factors in the final sampling decision during the sampling process. According to the assigned weights, the comprehensive reliability factor of each batch is calculated by multiplying the reliability factor of the first batch and the reliability factor of the second batch by their respective weights and then adding the results. Based on the comprehensive reliability factor, the sampling ratio of each batch is determined. The sampling ratio is used to allocate the number of samples to be drawn from each batch in the overall sample. According to the preset total sampling quantity, the actual sampling quantity of each batch is calculated. According to the calculated sampling quantity, samples are randomly drawn from the material set of each batch to form a conditional sampling sample set.

[0030] Furthermore, conditional sampling is performed on the K batches according to the K reliability factors of the first batch and the K reliability factors of the second batch to obtain K conditional sampling sample sets, including:

[0031] The K reliability factors of the first batch and the K reliability factors of the second batch are weighted and calculated to obtain K conditional sampling probability coefficients. Based on the magnitudes of the K conditional sampling probability coefficients, the areas of K roulette wheel partitions in the conditional sampling roulette are determined. The conditional sampling roulette is rotated multiple times according to the preset extraction quantity to obtain the K conditional sampling sample sets. Among them, each roulette wheel partition corresponds to one batch.

[0032] Preferably, the weights of the reliability factors of the first batch and the second batch are preset, and the K reliability factors of the first batch and the second batch are weighted to calculate the conditional sampling probability coefficient of each batch. According to the conditional sampling probabilities of the K batches, a conditional sampling roulette is designed. The roulette is divided into K regions, and each region corresponds to one batch. The area of the region is proportional to the conditional sampling probability of the batch. According to actual needs, the total number of samples to be drawn is set. According to the set extraction quantity, the conditional sampling roulette is rotated multiple times. After each rotation, the batch selected is determined according to the region where the pointer stops when the roulette stops. A sample is randomly drawn from the selected batch and added to the corresponding conditional sampling sample set. This process is repeated until the set extraction quantity is reached. After multiple rotations and extractions, the obtained samples are classified according to batches to form K conditional sampling sample sets.

[0033] The K conditional sampling sample sets are traversed for surface defect detection to obtain a set of surface defect detection results for the K conditional sampling samples.

[0034] Traverse the K sets of conditional sampling samples, and conduct surface defect detection on each sample in each set one by one. Specifically, use the detection equipment to identify the surface defects of each sample, and record the detection results of each sample, including the defect type (such as crack, air hole or scratch, etc.) and the defect degree (such as slight, severe, etc.); for each batch of conditional sampling samples, summarize all the detection results to form the surface defect detection result set of this batch of conditional sampling samples; by sequentially detecting the samples of K batches, finally obtain K surface defect detection result sets of conditional sampling samples.

[0035] Perform homogeneous aggregation on the K surface defect detection result sets of conditional sampling samples to obtain K clustering detection result clusters. Each clustering detection result cluster includes multiple clustering conditional sampling sample surface defect detection result sets, and each clustering conditional sampling sample surface defect detection result set includes multiple conditional sampling sample surface defect detection results belonging to the same surface defect type.

[0036] Classify the samples according to the defect type (such as crack, air hole, scratch, etc.) of each sample in the detection results, and group the samples belonging to the same defect type into one category. For each batch of detection result sets, extract the sample detection results with the same defect type to form multiple clustering conditional sampling sample surface defect detection result sets, and each set corresponds to a specific defect type. Further aggregate these homogeneous detection results to generate the clustering detection result cluster of each batch. Each cluster contains the clustering result sets of all different defect types detected in this batch. By sequentially processing the detection results of K batches, finally obtain K clustering detection result clusters.

[0037] Furthermore, performing homogeneous aggregation on the K surface defect detection result sets of conditional sampling samples to obtain K clustering detection result clusters includes:

[0038] Extract K first conditional sampling sample surface defect detection result sets from the K conditional sampling sample surface defect detection result sets respectively, and store them in K first aggregation nodes; perform binary classification on the K conditional sampling sample surface defect detection result sets according to a preset similarity threshold in the K first aggregation nodes, add the detection results whose similarity to the K first conditional sampling sample surface defect detection result sets meets the preset similarity threshold to the K first clustering conditional sampling sample surface defect detection result sets, and add the remaining detection results to the K first sets to be partitioned; extract K second conditional sampling sample surface defect detection result sets from the K first sets to be partitioned again, and store them in K second aggregation nodes; perform binary classification on the K first sets to be partitioned according to the preset similarity threshold in the K second aggregation nodes, add the detection results whose similarity to the K second conditional sampling sample surface defect detection result sets meets the preset similarity threshold to the K second clustering conditional sampling sample surface defect detection result sets, and add the remaining detection results to the K second sets to be partitioned; after multiple aggregation analyses until all the detection results in the K conditional sampling sample surface defect detection result sets are partitioned, obtain the K clustering detection result clusters.

[0039] Specifically, extract a representative detection result from the surface defect detection result set of the conditional sampling samples in each batch to form K first conditional sampling sample surface defect detection result sets, and store these sets in K first aggregation nodes respectively; perform binary classification on the detection result sets of each batch according to a preset similarity threshold: classify the detection results whose similarity to the detection result sets in the first aggregation nodes meets the threshold into the K first clustering conditional sampling sample surface defect detection result sets; the remaining detection results are assigned to the K first sets to be partitioned. For the K first sets to be partitioned, extract a new representative detection result respectively to form K second conditional sampling sample surface defect detection result sets, and store them in K second aggregation nodes; in a similar manner, perform binary classification on the sets to be partitioned: classify the detection results that meet the similarity threshold into the K second clustering conditional sampling sample surface defect detection result sets; the remaining detection results are assigned to the K second sets to be partitioned. According to the above steps, successively extract, store in new aggregation nodes, classify and partition the remaining detection results in the sets to be partitioned; as the number of iterations increases, the unpartitioned detection results gradually decrease, and the clustering conditional sample sets gradually expand; continue multiple rounds of aggregation analysis until all detection results are classified into the corresponding clustering conditional sampling sample sets and the sets to be partitioned are empty; finally, form K clustering detection result clusters, and each cluster contains the complete set of all detection results in this batch aggregated according to the defect type.

[0040] Identify the defect eigenvalue for the K cluster detection result clusters, and conduct a centralized analysis of the identification results to determine the set of K cluster defect eigenvalue central values.

[0041] Extract the features related to surface defects from each cluster detection result cluster, such as the size, shape, location, color, etc. of the defects; use machine learning algorithms or statistical methods to quantify the extracted features to obtain the specific values of each feature, and these values will be used as defect eigenvalues. For the set of eigenvalues in each cluster detection result cluster, calculate the centralized statistical indicators of the eigenvalues, such as the mean, variance, extreme values, etc., to comprehensively reflect the overall distribution of the sample features in this cluster; after completing the centralized analysis of all the eigenvalues of each cluster detection result cluster, form a set of K cluster defect eigenvalue central values, and each set includes the centralized value statistical results of all the features in this cluster.

[0042] Furthermore, it includes:

[0043] Extract the set of surface defect detection results of K first-cluster conditional sampling samples from the K cluster detection result clusters; use a semantic analyzer to identify the defect eigenvalues for the set of surface defect detection results of the K first-cluster conditional sampling samples to obtain a set of K first-cluster defect eigenvalues; traverse the set of K first-cluster defect eigenvalues to conduct a centralized analysis of the eigenvalues to determine the K first-cluster defect eigenvalue central values; conduct a centralized analysis of the defect eigenvalues for the K cluster detection result clusters to obtain the set of K cluster defect eigenvalue central values.

[0044] Specifically, select a certain number of representative conditional sampling sample surface defect detection result sets from each cluster detection result cluster as the set of surface defect detection results of the first-cluster conditional sampling samples; preprocess the extracted set of surface defect detection results of the first-cluster conditional sampling samples, such as data cleaning, formatting, etc., to ensure the accuracy and consistency of the data; use a semantic analyzer to identify the defect eigenvalues for each set of surface defect detection results of the first-cluster conditional sampling samples, and extract the key eigenvalues related to surface defects. For example, the semantic analyzer can automatically identify the geometric dimensions, location distribution, morphological features, etc. of the defects based on specific rules or models to form a set of K first-cluster defect eigenvalues; traverse each set of first-cluster defect eigenvalues, conduct a statistical analysis of all the eigenvalues in the set, and calculate the centralized indicators of the eigenvalues, such as the mean, variance, or median, etc., to determine the K first-cluster defect eigenvalue central values; based on the results of the first-step centralized analysis of the eigenvalues, further conduct a more extensive analysis of the defect eigenvalues for the K cluster detection result clusters, covering the eigenvalues of all samples in each cluster, calculate the set of feature central values of all samples, including the comprehensive statistical indicators of all features, and finally form a set of K cluster defect eigenvalue central values.

[0045] Furthermore, it includes:

[0046] Calculate the mean of the K sets of first clustering defect eigenvalue to obtain K first clustering defect feature means; use the K first clustering defect feature means as indexes, and iterate in the K sets of first clustering defect eigenvalues according to a preset iteration step length to obtain K first iterative clustering defect eigenvalues; determine whether the concentration measure of the K first iterative clustering defect eigenvalues is greater than or equal to the concentration measure of the K first clustering defect feature means. If so, use the K first iterative clustering defect eigenvalues as indexes and iterate in the K sets of first clustering defect eigenvalues according to the preset iteration step length until the preset number of iterations is satisfied, to obtain K first target clustering defect eigenvalues, and use the K first target clustering defect eigenvalues as the central values of the K first clustering defect feature sets.

[0047] Calculate the mean of the defect eigenvalue sets of each cluster, that is, calculate the mean of all eigenvalues in the K sets of first clustering defect eigenvalues to obtain K first clustering defect feature means; use the K first clustering defect feature means as indexes and perform iterative update in the defect eigenvalue sets of each cluster according to a preset iteration step length; through each round of iteration, the eigenvalue set will gradually approach the true features of the cluster, so as to obtain K first iterative clustering defect eigenvalues; compare the K first iterative clustering defect eigenvalues with their corresponding means to determine whether their concentration measure is greater than or equal to the concentration measure of the initial means; if not, stop the iteration, and use the K first clustering defect feature means as the set of central values of the K first clustering defect feature sets; if the concentration measure of the K first iterative clustering defect eigenvalues meets the conditions, use these iterative values as new indexes and iterate again in the eigenvalue set according to the same iteration step length; when the iteration process reaches the preset number of iterations and the concentration measure of the eigenvalue set meets the stability condition, finally obtain K first target clustering defect eigenvalues, and these values are used as the final central values of the features of each cluster; use the K first target clustering defect eigenvalues as the final central values of the first clustering defect feature sets.

[0048] Furthermore, count the number of first clustering defect eigenvalues in the K sets of first clustering defect eigenvalues whose distance to the K first clustering defect feature means is the preset iteration step length, to obtain the concentration measure of the K first iterative clustering defect eigenvalues.

[0049] For each sample in the first set of clustering defect eigenvalue sets, calculate the distance from the sample to the corresponding clustering feature mean. The Euclidean distance or Manhattan distance can be used to measure this distance, which represents the degree of difference between each eigenvalue and its corresponding mean. After calculating the distance from each sample to the mean, count the number of samples whose distances are less than or equal to the preset iteration step size. This number indicates how many samples in the eigenvalue set have changes within the preset step size range in one iteration. Use the counted number of qualified samples as the concentration measure of the clustering. This concentration measure reflects the stability and consistency of the samples in the current iteration. The larger the concentration measure, the more concentrated the eigenvalues in the current clustering, and the more stable the clustering result. The concentration measure can usually be expressed as a proportion or an absolute value of this number.

[0050] Perform cross-batch weighted calculation on the K sets of clustering defect feature central values according to the preset weights to obtain a weighted set of clustering defect feature central values, and use the weighted set of clustering defect feature central values as the K surface defect detection results.

[0051] Perform weighted calculation on the K sets of clustering defect feature central values according to the preset weights. First, set a weight for each batch. These weights can be determined according to the importance of the batch or other relevant factors. Then, weight the defect feature central values of each batch according to these weights. The specific method is to multiply the feature central value of each batch by its corresponding weight and sum to obtain the weighted set of clustering defect feature central values. Finally, use the obtained weighted set of clustering defect feature central values as the K surface defect detection results. Through this weighted calculation, the defect features of different batches can be combined and adjusted according to the importance of each batch, so as to more accurately reflect the defect situation and quality level of the overall material.

[0052] In summary, the embodiments of the present application have at least the following technical effects:

[0053] First, obtain the K batches and the quantities of the K batches of the target building materials, where K is an integer greater than or equal to 1. Then, respectively conduct initial sample extraction of a preset extraction quantity on the K batches in a random sampling manner to obtain K initial extraction sample sets, and conduct surface defect detection on the K initial extraction sample sets, and obtain K first-batch reliability factors according to the detection results; at the same time, conduct volume analysis based on the quantities of the K batches to determine K second-batch reliability factors. Further, conduct conditional sampling on the K batches according to the K first-batch reliability factors and the K second-batch reliability factors to obtain K conditional sampling sample sets; and traverse the K conditional sampling sample sets to conduct surface defect detection to obtain a set of surface defect detection results of the K conditional sampling samples. Then, conduct homogeneous aggregation on the set of surface defect detection results of the K conditional sampling samples to obtain K clustering detection result clusters, each clustering detection result cluster includes multiple clustering conditional sampling sample surface defect detection result sets, and each clustering conditional sampling sample surface defect detection result set includes multiple conditional sampling sample surface defect detection results belonging to the same surface defect type. Next, identify the defect characteristic values of the K clustering detection result clusters, and conduct centralized analysis of the identification results to determine K clustering defect characteristic centralized value sets. Finally, perform cross-batch weighted calculation on the K clustering defect characteristic centralized value sets according to a preset weight to obtain a weighted clustering defect characteristic centralized value set, and use the weighted clustering defect characteristic centralized value set as the surface defect detection results of the K batches. This solves the technical problem of low efficiency and accuracy of surface defect detection of building materials in the prior art, and achieves the technical effect of improving detection efficiency and accuracy.

[0054] Embodiment 2, based on the same inventive concept as the method for intelligent detection of surface defects of building materials in the foregoing embodiment, as Figure 2 shown, the present application provides a system for intelligent detection of surface defects of building materials, wherein the system includes:

[0055] A data acquisition module 11, which is used to acquire K batches and the quantities of K batches of target building materials, where K is an integer greater than or equal to 1; a first sampling module 12, which is used to respectively extract initial samples with a preset extraction quantity from the K batches in a random sampling manner to obtain K initial extraction sample sets, and perform surface defect detection on the K initial extraction sample sets, and obtain K first batch reliability factors according to the detection results; an analysis module 13, which is used to perform volume analysis based on the quantities of the K batches to determine K second batch reliability factors; a second sampling module 14, which is used to perform conditional sampling on the K batches according to the K first batch reliability factors and the K second batch reliability factors to obtain K conditional sampling sample sets; a defect detection module 15, which is used to traverse the K conditional sampling sample sets to perform surface defect detection to obtain a set of surface defect detection results of the K conditional sampling samples; an aggregation module 16, which is used to perform homogeneous aggregation on the set of surface defect detection results of the K conditional sampling samples to obtain K clustering detection result clusters, each clustering detection result cluster includes multiple clustering conditional sampling sample surface defect detection result sets, and each clustering conditional sampling sample surface defect detection result set includes multiple conditional sampling sample surface defect detection results belonging to the same surface defect type; an identification module 17, which is used to identify defect characteristic values of the K clustering detection result clusters, and perform centralized analysis of the identification results to determine K clustering defect characteristic centralized value sets; a calculation module 18, which is used to perform cross-batch weighted calculation on the K clustering defect characteristic centralized value sets according to a preset weight to obtain a weighted clustering defect characteristic centralized value set, and use the weighted clustering defect characteristic centralized value set as the surface defect detection results of the K batches.

[0056] Further, the second sampling module 14 is used to execute the following method:

[0057] Perform weighted calculation on the K first batch reliability factors and the K second batch reliability factors to obtain K conditional sampling probability coefficients; determine the areas of K roulette wheel partitions in the conditional sampling roulette according to the magnitudes of the K conditional sampling probability coefficients, and rotate the conditional sampling roulette multiple times according to the preset extraction quantity to obtain the K conditional sampling sample sets; where each roulette wheel partition corresponds to one batch.

[0058] Further, the identification module 17 is used to execute the following method:

[0059] Extract K sets of surface defect detection results of the first clustering conditional sampling samples from the K clusters of clustering detection results; use a semantic analyzer to identify defect eigenvalue of the K sets of surface defect detection results of the first clustering conditional sampling samples to obtain K sets of the first clustering defect eigenvalues; traverse the K sets of the first clustering defect eigenvalues for eigenvalue concentration analysis to determine K central values of the first clustering defects; perform eigenvalue concentration analysis on the K clusters of clustering detection results to obtain the set of K central values of the clustering defect eigenvalues.

[0060] Further, the recognition module 17 is used to execute the following method:

[0061] Calculate the mean of the K sets of the first clustering defect eigenvalues to obtain K means of the first clustering defect eigenvalues; use the K means of the first clustering defect eigenvalues as indexes, and iterate in the K sets of the first clustering defect eigenvalues according to a preset iteration step length to obtain K first iterative clustering defect eigenvalues; determine whether the concentration of the K first iterative clustering defect eigenvalues is greater than or equal to the concentration of the K means of the first clustering defect eigenvalues. If so, use the K first iterative clustering defect eigenvalues as indexes, and iterate in the K sets of the first clustering defect eigenvalues according to the preset iteration step length until the preset iteration times are satisfied to obtain K first target clustering defect eigenvalues, and use the K first target clustering defect eigenvalues as the K central values of the first clustering defect eigenvalues.

[0062] Further, the recognition module 17 is used to execute the following method:

[0063] Count the number of the first clustering defect eigenvalues whose distance to the K means of the first clustering defect eigenvalues in the K sets of the first clustering defect eigenvalues is the preset iteration step length to obtain the concentration of the K first iterative clustering defect eigenvalues.

[0064] Further, the aggregation module 16 is used to execute the following method:

[0065] Extract K first conditional sampling sample surface defect detection result sets from the K conditional sampling sample surface defect detection result sets respectively, and store them in K first aggregation nodes; perform binary classification on the K conditional sampling sample surface defect detection result sets according to a preset similarity threshold in the K first aggregation nodes, add the detection results whose similarity to the K first conditional sampling sample surface defect detection result sets meets the preset similarity threshold into K first clustering conditional sampling sample surface defect detection result sets, and add the remaining detection results into K first sets to be partitioned; extract K second conditional sampling sample surface defect detection result sets from the K first sets to be partitioned again, and store them in K second aggregation nodes; perform binary classification on the K first sets to be partitioned according to a preset similarity threshold in the K second aggregation nodes, add the detection results whose similarity to the K second conditional sampling sample surface defect detection result sets meets the preset similarity threshold into K second clustering conditional sampling sample surface defect detection result sets, and add the remaining detection results into K second sets to be partitioned; through multiple aggregation analyses until all the detection results in the K conditional sampling sample surface defect detection result sets are partitioned, obtain the K clustering detection result clusters.

[0066] Further, the first sampling module 12 is used to execute the following method:

[0067] Perform surface defect detection on the K initial extraction sample sets to obtain K initial extraction sample detection result sets; traverse the K initial extraction sample detection result sets for result stability analysis to obtain K stability factors; respectively use the ratios of the K stability factors to the sum of the K stability factors as the K first batch reliability factors.

[0068] It should be noted that the above sequence of embodiments of the present application is only for description and does not represent the advantages or disadvantages of the embodiments. And the above describes specific embodiments of this specification. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0069] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0070] This specification and the drawings are merely illustrative of the present application and are considered to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to cover these changes and modifications.

Claims

1. An intelligent detection method for surface defects of building materials, characterized in that: The method comprises: Obtaining K batches and K batch quantities of target building materials, where K is an integer greater than or equal to 1; Extracting a preset number of initial samples from the K batches in a random sampling manner to obtain K initial sample sets, and performing surface defect detection on the K initial sample sets to obtain K first batch reliability factors according to the detection results; Performing volume analysis based on the K batch numbers to determine K second batch reliability factors; Perform conditional sampling on the K batches according to the K first batch reliability factors and the K second batch reliability factors to obtain K conditional sampling sample sets; Traversing the K conditional sampling sample sets to perform surface defect detection, and obtaining a surface defect detection result set of the K conditional sampling samples; Performing similar aggregation on the K conditional sampling sample surface defect detection result sets to obtain K cluster detection result clusters, each cluster detection result cluster includes multiple cluster conditional sampling sample surface defect detection result sets, and each cluster conditional sampling sample surface defect detection result set includes multiple conditional sampling sample surface defect detection results belonging to the same surface defect type; Perform defect feature value identification on the K cluster detection result clusters, and perform feature value centralized analysis on the identification results to determine a set of K cluster defect feature centralized values; The K clustered defect feature concentrated value sets are weighted calculated across batches according to preset weights to obtain a weighted clustered defect feature concentrated value set, and the weighted clustered defect feature concentrated value set is used as the K surface defect detection results.

2. The intelligent detection method for surface defects of building materials according to claim 1, characterized in that: Conditionally sampling the K batches according to the K first batch reliability factors and the K second batch reliability factors to obtain K conditional sampling sample sets, including: Performing weighted calculation on the K first batch reliability factors and the K second batch reliability factors to obtain K conditional sampling probability coefficients; The areas of K roulette wheel partitions in the conditional sampling roulette are determined based on the sizes of the K conditional sampling probability coefficients, and the conditional sampling roulette wheel is turned multiple times according to a preset number of draws to obtain the K conditional sampling sample sets; wherein each roulette wheel partition corresponds to a batch.

3. The intelligent detection method for surface defects of building materials according to claim 1, characterized in that: include: Extracting K first cluster conditional sampling sample surface defect detection result sets from the K cluster detection result clusters; Using a semantic analyzer to identify defect feature values ​​on the K first cluster conditional sampling sample surface defect detection result sets, to obtain K first cluster defect feature value sets; Traversing the K first cluster defect feature value sets to perform feature value concentration analysis, and determining K first cluster defect feature concentration values; Perform defect feature value concentration analysis on the K cluster detection result clusters to obtain the K cluster defect feature concentration value sets.

4. The intelligent detection method for surface defects of building materials according to claim 3, characterized in that: include: Calculating the mean of the K first cluster defect feature value sets to obtain the K first cluster defect feature means; Taking the mean values ​​of the K first clustering defect features as indexes, iterating in the K first clustering defect feature value sets according to a preset iteration step length to obtain K first iteration clustering defect feature values; Determine whether the concentration of the K first iterative clustering defect feature values ​​is greater than or equal to the concentration of the K first clustering defect feature mean values. If so, use the K first iterative clustering defect feature values ​​as indexes, and iterate in the K first clustering defect feature value sets according to a preset iteration step until a preset number of iterations is met, so as to obtain K first target clustering defect feature values, and use the K first target clustering defect feature values ​​as the concentration values ​​of the K first clustering defect features.

5. The intelligent detection method for surface defects of building materials according to claim 4, characterized in that: The number of first cluster defect feature values ​​whose distance from the K first cluster defect feature means is the preset iteration step in the K first cluster defect feature value sets is counted to obtain the concentration of the K first iteration cluster defect feature values.

6. The intelligent detection method for surface defects of building materials according to claim 1, characterized in that: The K conditional sampling sample surface defect detection result sets are aggregated into the same category to obtain K cluster detection result clusters, including: Extracting K first conditional sampling sample surface defect detection result sets from the K conditional sampling sample surface defect detection result sets respectively, and storing them in K first aggregation nodes; Binary classification is performed on the K conditional sampling sample surface defect detection result sets according to a preset similarity threshold at the K first aggregation nodes, and detection results whose similarity with the K first conditional sampling sample surface defect detection result sets meets the preset similarity threshold are added into the K first clustering conditional sampling sample surface defect detection result sets, and the remaining detection results are added into the K first to-be-divided sets; Extracting K second conditional sampling sample surface defect detection result sets from the K first to-be-divided sets again, and storing them in K second aggregation nodes; Binary classification is performed on the K first to-be-divided sets at the K second aggregation nodes according to a preset similarity threshold, and the detection results whose similarity with the K second conditional sampling sample surface defect detection result sets meets the preset similarity threshold are added into the K second clustering conditional sampling sample surface defect detection result sets, and the remaining detection results are added into the K second to-be-divided sets; After multiple aggregation analyses, until all the detection results in the K conditional sampling sample surface defect detection result sets are divided, the K cluster detection result clusters are obtained.

7. The intelligent detection method for surface defects of building materials according to claim 1, characterized in that: Obtain K initial sample sets, perform surface defect detection on the K initial sample sets, and obtain K first batch reliability factors according to the detection results, including: Performing surface defect detection on the K initial sample sets to obtain K initial sample detection result sets; Traversing the K initial sample test result sets to perform result stability analysis and obtain K stability factors; The ratios of the K stability factors to the sum of the K stability factors are respectively used as the K first batch reliability factors.

8. Intelligent detection system for building material surface defects, characterized in that: For implementing the intelligent detection method for surface defects of building materials according to any one of claims 1 to 7, the system comprises: A data acquisition module, the data acquisition module is used to obtain K batches and K batch quantities of target building materials, wherein K is an integer greater than or equal to 1; A first sampling module, the first sampling module is used to extract a preset number of initial samples from the K batches in a random sampling manner to obtain K initial sample sets, and perform surface defect detection on the K initial sample sets to obtain K first batch reliability factors according to the detection results; An analysis module, the analysis module is used to perform volume analysis based on the K batch quantities to determine K second batch reliability factors; A second sampling module, the second sampling module is used to perform conditional sampling on the K batches according to the K first batch reliability factors and the K second batch reliability factors to obtain K conditional sampling sample sets; A defect detection module, the defect detection module is used to traverse the K conditional sampling sample sets to perform surface defect detection and obtain a surface defect detection result set of the K conditional sampling samples; An aggregation module, wherein the aggregation module is used to perform similar aggregation on the K conditional sampling sample surface defect detection result sets to obtain K cluster detection result clusters, each cluster detection result cluster includes multiple cluster conditional sampling sample surface defect detection result sets, and each cluster conditional sampling sample surface defect detection result set includes multiple conditional sampling sample surface defect detection results belonging to the same surface defect type; An identification module, the identification module is used to identify defect feature values ​​for the K cluster detection result clusters, and perform feature value centralized analysis on the identification results to determine a set of K cluster defect feature centralized values; A calculation module is used to perform cross-batch weighted calculation on the K clustered defect feature concentrated value sets according to preset weights to obtain a weighted clustered defect feature concentrated value set, and use the weighted clustered defect feature concentrated value set as the K surface defect detection results.

Images