A data processing method and system for bridge crack detection

Abstract

The present invention relates to the field of crack data processing technology, and discloses a data processing method and system for bridge crack detection, the method comprising: deploying multiple data detection areas for bridge cracks to be detected, determining target crack length data according to preprocessing results, aggregating a first set of cracks and a second set of cracks using a clustering algorithm to obtain a first cluster set and a second cluster set respectively, and calculating a crack length factor according to the first data and the second data; comparing the crack length factor with the historical crack length factor of the bridge crack, judging whether to correct the crack length factor and obtain the target crack length factor according to the comparison result, setting a correction constant of the target crack length factor according to a first correction number, a second correction number and a third correction number, and obtaining final crack length data. The present invention dynamically optimizes the initial crack length data to ensure the authenticity and reliability of the final crack length data.

Description

Technical Field

[0001] The present invention relates to the technical field of crack data processing, and in particular to a data processing method and system for bridge crack detection. Background Art

[0002] As bridges age and their service life increases, cracks become one of the most common forms of damage in bridge structures. Crack length monitoring has become an important part of bridge health management. However, with the complexity of bridge structures and the diversity of crack occurrence patterns, traditional crack detection methods have gradually exposed problems such as insufficient accuracy, complex operation, and low efficiency. At present, crack detection mainly relies on two methods: manual detection and sensor-based automatic detection. Manual detection measures the length and width of cracks through visual observation or using simple measuring tools. Although intuitive data can be obtained, due to subjective factors of the inspectors, improper sensor installation position, inaccurate equipment calibration, or aging of the equipment itself, the length data provided itself has errors, and the data volume is huge, which often requires further processing and analysis. The accuracy of data processing directly affects the reliability of crack length detection results.

[0003] Therefore, how to provide a data processing method and system for bridge crack detection is a technical problem that technical personnel in this field urgently need to solve. Summary of the invention

[0004] In view of this, the present invention proposes a data processing method and system for bridge crack detection, aiming to solve the problem that the length data currently provided has errors and the data volume is huge, making it impossible to accurately obtain crack length detection results.

[0005] In one aspect, the present invention provides a data processing method for bridge crack detection, comprising:

[0006] Deploy multiple data detection areas for the bridge cracks to be detected, obtain initial crack length data corresponding to each data detection area, pre-process the initial crack length data, and determine target crack length data according to the pre-processing results;

[0007] The target crack length data is divided into first length data and second length data according to a set length threshold, and the first length data and second length data of the same type are extracted from each data detection area, and are respectively optimized to construct a first crack set and a second crack set;

[0008] The first set of cracks and the second set of cracks are aggregated by using a clustering algorithm to obtain a first cluster set and a second cluster set respectively, first data and second data are obtained according to the first cluster set and the second cluster set respectively, and a crack length factor is calculated according to the first data and the second data;

[0009] Comparing the crack length factor with the historical crack length factor of the bridge crack, and judging whether to correct the crack length factor and obtain a target crack length factor according to the comparison result;

[0010] Establishing a length data correction factor for the corresponding data detection area according to the first length data and the second length data, wherein the length data correction factor includes a first correction factor, a second correction factor and a third correction factor;

[0011] Counting a first correction quantity of the first correction factor, counting a second correction quantity of the second correction factor, counting a third correction quantity of the third correction factor, and setting a correction constant of the target crack length factor according to the first correction quantity, the second correction quantity, and the third correction quantity;

[0012] The target crack length factor is corrected according to the correction constant to obtain final crack length data.

[0013] Further, when the target crack length data is divided into first length data and second length data according to a set length threshold, it includes:

[0014] The comprehensive score of bridge cracks is calculated based on the bridge service factor, bridge material factor, design standard factor and bridge maintenance factor. The comprehensive score of bridge cracks is calculated by the following formula:

[0015] ;

[0016] ;

[0017]

[0018] in, represents the bridge service factor, Indicates the current year, Indicates the year the bridge was built. represents the design service life of the bridge, represents the bridge maintenance factor, Indicates the actual number of maintenance times. Indicates the recommended maintenance times for the bridge. represents the bridge material factor, represents the design standard factor, , , and is the weight coefficient, and , It represents the comprehensive score of bridge cracks;

[0019] The length threshold is determined according to the comprehensive score of the bridge cracks, and is divided into first length data and second length data according to the set length threshold.

[0020] Further, when the length threshold is determined according to the comprehensive score of the bridge crack, and the length threshold is divided into the first length data and the second length data according to the set length threshold, it includes:

[0021] when When it is greater than or equal to 0.8, it is set as the first length threshold;

[0022] when When it is less than 0.8, it is set as the second length threshold;

[0023] Classify the target crack length data that is greater than or equal to the length threshold as the first length data;

[0024] The target crack length data that is smaller than the length threshold is classified as the second length data.

[0025] Furthermore, when the first set of fractures and the second set of fractures are aggregated by using a clustering algorithm to obtain a first cluster set and a second cluster set respectively, the method includes:

[0026] Set the number of clusters k=2, randomly select two first length data in the first set of cracks as the first cluster center and the second cluster center, calculate the distance to the first cluster center and the second cluster center for each first length data in the first set of cracks, and assign it to the cluster with the closest distance. If the distances are the same, assign it to the cluster where the first cluster center is located.

[0027] The distances of each first length data in the first set of cracks to the center of the first cluster and the center of the second cluster are calculated respectively, and are obtained by the following formula:

[0028] ;

[0029] ;

[0030] in, represents any first length data in the first set of cracks, represents the first cluster center, represents the second cluster center, represents the distance from any first-length data in the first set to the center of the first cluster, Indicates the distance from any first-length data in the first set to the center of the second cluster;

[0031] According to the allocation result, the first cluster set and the second cluster set are obtained.

[0032] Further, when first data and second data are obtained respectively according to the first cluster set and the second cluster set, and a crack length factor is calculated according to the first data and the second data, it includes:

[0033] Taking an average value of the cluster where the center of the first cluster is located in the first cluster set to obtain a first mean, taking an average value of the cluster where the center of the second cluster is located in the first cluster set to obtain a second mean, and obtaining a third mean and a fourth mean based on the second cluster set;

[0034] The crack length factor is calculated based on the first mean, the second mean, the third mean and the fourth mean. The crack length factor is obtained by the following formula:

[0035]

[0036] in, represents the crack length factor, represents the first mean, represents the second mean, represents the third mean, represents the fourth mean, , , and is the weight coefficient, and .

[0037] Furthermore, when comparing the crack length factor with the historical crack length factor of the bridge crack, judging whether to correct the crack length factor and deriving the target crack length factor according to the comparison result, it includes:

[0038] When the crack length factor is greater than or equal to the minimum value of the historical crack length factors, it is determined that the crack length factor is not to be corrected, and the crack length factor is determined as a target crack length factor;

[0039] When the crack length factor is less than the minimum value of the historical crack length factor, determining to correct the crack length factor;

[0040] When it is determined to correct the crack length factor, the crack length factor is compared with the average of the historical crack length factors, and a final correction factor is determined according to the factor similarity.

[0041] Furthermore, when comparing the crack length factor with the average value of the historical crack length factor and determining the final correction factor according to the factor similarity, it includes:

[0042] The factor similarity is obtained by the following formula:

[0043]

[0044]

[0045] in, represents the standard deviation of the historical crack length factor, represents the number of historical crack length factors, represents the i-th historical crack length factor, is the mean value of the historical crack length factor, represents the crack length factor, Indicates factor similarity;

[0046] when When it is greater than or equal to 0.7, the first preset correction factor is used as the final correction factor of the crack length factor;

[0047] when When it is less than 0.7 and greater than or equal to 0, the second preset correction factor is used as the final correction factor of the crack length factor;

[0048] when When it is less than 0, the third preset correction factor is used as the final correction factor of the crack length factor;

[0049] The target crack length factor is determined according to the product value of the final correction factor and the crack length factor.

[0050] Further, when establishing a length data correction factor for the corresponding data detection area according to the first length data and the second length data, it includes:

[0051] Determine the first standard crack length data and the second standard crack length data corresponding to each data detection area;

[0052] comparing the first length data and the second length data with the first standard crack length data and the second standard crack length data respectively;

[0053] When all the first length data are less than or equal to the first standard crack length data and all the second length data are less than or equal to the second standard crack length data, determining a corresponding data detection area to generate the first correction factor;

[0054] When all the first length data are greater than the first standard crack length data and all the second length data are greater than the second standard crack length data, determining the corresponding data detection area to generate the second correction factor;

[0055] When there are one or more first length data that are not equal to the first standard crack length data and there are one or more second length data that are not equal to the second standard crack length data, the corresponding data detection area is determined to generate the third correction factor.

[0056] Further, when the correction constant of the target crack length factor is set according to the first correction amount, the second correction amount and the third correction amount, it includes:

[0057] Determine a first correction number, a second correction number, and a third correction number, perform a normalization process on them, and obtain a normalized value J;

[0058] The preset calibration constants are a first preset calibration constant, a second preset calibration constant and a third preset calibration constant;

[0059] When J≤2, the first preset correction constant is used as the correction constant of the target crack length factor;

[0060] When 2<J≤3.5, the second preset correction constant is used as the correction constant of the target crack length factor;

[0061] When 3.5<J, the third preset correction constant is used as the correction constant of the target crack length factor;

[0062] The final crack length data is determined according to the product value of the calibration constant and the target crack length.

[0063] Compared with the prior art, the beneficial effects of the present invention are as follows: by preprocessing and optimizing the initial crack length data, the error in the measurement process can be effectively removed, the accuracy of the target crack length data obtained can be ensured, and the influence of accidental errors and human factors on the accuracy of the crack length data can be reduced, thereby improving the accuracy of crack detection. By clustering the target crack length data and distinguishing the crack data of different lengths, redundant data can be effectively eliminated, the availability and reliability of the data can be improved, and the crack length factor can be compared with the historical crack length factor to find abnormal fluctuations in the data, and on this basis, the crack length factor can be corrected. The accuracy of the crack detection results can be further improved, and the result errors caused by single detection errors can be avoided. By establishing a length data correction factor, the influence of data deviation on the final crack length data can be avoided, the refinement of the correction process can be improved, the length data of the bridge cracks can be accurately reflected, and the risk of bridge accidents can be reduced.

[0064] On the other hand, the present application also provides a data processing system for bridge crack detection, which is used to apply the above-mentioned data processing method for bridge crack detection, including:

[0065] Acquisition unit: deploy multiple data detection areas for the bridge cracks to be detected, obtain initial crack length data corresponding to each data detection area, pre-process the initial crack length data, and determine target crack length data according to the pre-processing result;

[0066] Aggregation unit: divides the target crack length data into first length data and second length data according to a set length threshold, extracts the same type of first length data and second length data from each data detection area, performs optimization processing on them respectively, and constructs a first crack set and a second crack set;

[0067] The first set of cracks and the second set of cracks are aggregated by using a clustering algorithm to obtain a first cluster set and a second cluster set respectively, first data and second data are obtained according to the first cluster set and the second cluster set respectively, and a crack length factor is calculated according to the first data and the second data;

[0068] A first correction unit: compares the crack length factor with the historical crack length factor of the bridge crack, determines whether to correct the crack length factor according to the comparison result, and obtains a target crack length factor;

[0069] A second correction unit: establishing a length data correction factor for a corresponding data detection area according to the first length data and the second length data, wherein the length data correction factor includes a first correction factor, a second correction factor and a third correction factor;

[0070] A length data confirmation unit: counting a first correction number of the first correction factor, counting a second correction number of the second correction factor, counting a third correction number of the third correction factor, and setting a correction constant of the target crack length factor according to the first correction number, the second correction number, and the third correction number;

[0071] The target crack length factor is corrected according to the correction constant to obtain final crack length data.

[0072] It can be understood that the above-mentioned data processing method and system for bridge crack detection have the same beneficial effects, which will not be described in detail here. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0074] Figure 1 A flow chart of a data processing method for bridge crack detection provided by an embodiment of the present invention;

[0075] Figure 2 A structural block diagram of a data processing system for bridge crack detection provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0076] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0077] In some embodiments of the present application, see Figure 1 As shown, a data processing method for bridge crack detection includes:

[0078] S100: deploying multiple data detection areas for the bridge cracks to be detected, obtaining initial crack length data corresponding to each data detection area, preprocessing the initial crack length data, and determining target crack length data according to the preprocessing result;

[0079] S200: dividing the target crack length data into first length data and second length data according to a set length threshold, extracting the same type of first length data and second length data from each data detection area, respectively performing optimization processing and constructing a first crack set and a second crack set;

[0080] S300: clustering the first set of cracks and the second set of cracks using a clustering algorithm to obtain a first cluster set and a second cluster set respectively, obtaining first data and second data respectively according to the first cluster set and the second cluster set, and calculating a crack length factor according to the first data and the second data;

[0081] S400: comparing the crack length factor with the historical crack length factor of the bridge crack, and determining whether to correct the crack length factor according to the comparison result and obtaining a target crack length factor;

[0082] S500: Establishing a length data correction factor for a corresponding data detection area according to the first length data and the second length data, wherein the length data correction factor includes a first correction factor, a second correction factor and a third correction factor;

[0083] S600: Counting a first correction number of a first correction factor, counting a second correction number of a second correction factor, counting a third correction number of a third correction factor, and setting a correction constant of a target crack length factor according to the first correction number, the second correction number, and the third correction number;

[0084] S700: Correcting the target crack length factor according to the correction constant to obtain final crack length data.

[0085] Specifically, multiple data detection areas are deployed at the crack locations of the bridge. The data detection areas are arranged according to the structure of the bridge and the locations where the cracks appear to ensure that all cracks at the crack locations are covered. The corresponding initial crack length data of each data detection area is obtained, and the initial crack length data is preprocessed. The preprocessing includes removing erroneous data and environmental noise of the initial crack length data, thereby improving the reliability and clarity of the initial crack length data. After preprocessing, the target crack length data is obtained. The target crack length data is divided into first length data and second length data according to a set length threshold. These two types of data represent different degrees of crack length. The first length data represents a longer crack length, while the second length data represents a shorter crack length. For each data detection area, the first length data and the second length data of the same type are extracted respectively. The same type means that the directions of the cracks are consistent, for example: all are longitudinal cracks, or all are transverse cracks, or all are oblique cracks. The optimization process includes removing cracks with wrong directions and eliminating redundant cracks, ensuring the accuracy of the first length data and the second length data of the same type, respectively summarizing the first length data and the second length data to construct the first set of cracks and the second set of cracks, and using the clustering algorithm for aggregation processing to summarize the potential laws of the crack length, and obtain the first cluster set and the second cluster set of cracks. The crack length factor is calculated based on the first data and the second data obtained from the clustering results. The crack length factor is a comprehensive length value of the crack length, which reflects the actual length of the bridge crack. The crack length factor is compared with the historical crack length factor of the bridge. By comparing with the historical crack length factor, it can be judged whether the current crack length factor is reasonable. If the current calculation result has a large deviation, it needs to be corrected to obtain the target crack length factor. Correcting the crack length factor ensures the scientificity and reliability of the final crack length data. The correction constant obtained according to the length data correction factor reflects the deviation between the first data and the second data and the standard. The correction constant is used to finely correct the target crack length factor, further improving the rigor and accuracy of the final crack length data.

[0086] It can be understood that the final crack length data obtained by collecting, preprocessing, optimizing, clustering and correcting the initial crack length data avoids measurement errors and deviations in the detection process and improves the accuracy and reliability of the final crack length data.

[0087] In some embodiments of the present application, when the target crack length data is divided into the first length data and the second length data according to the set length threshold, it includes: calculating the comprehensive score of the bridge cracks according to the bridge service factor, the bridge material factor, the design standard factor, and the bridge maintenance factor, and the comprehensive score of the bridge cracks is obtained by the following formula:

[0088] ;

[0089] ;

[0090]

[0091] in, represents the bridge service factor, Indicates the current year, Indicates the year the bridge was built. represents the design service life of the bridge, represents the bridge maintenance factor, Indicates the actual number of maintenance times. Indicates the recommended maintenance times for the bridge. represents the bridge material factor, represents the design standard factor, , , and is the weight coefficient, and , It represents the comprehensive score of bridge cracks, determines the length threshold according to the comprehensive score of bridge cracks, and is divided into first length data and second length data according to the set length threshold.

[0092] It is understandable that in the process of analyzing and processing the target crack length data, the length threshold is set based on the comprehensive score of bridge cracks, which measures multiple factors affecting the occurrence of cracks in bridges. The bridge service factor reflects the influence of the service life of the bridge. The longer the service life, the higher the risk of longer cracks. The bridge material factor takes into account the type of material used in the bridge, and is set at 0.8 for concrete bridges, 1.0 for steel bridges, and 0.6 for mixed material bridges. The performance of bridge materials affects the occurrence and extension length of cracks. The design standard factor is the specification standard used in bridge design. The bridge built in the last 20 years is 0.3, the bridge built in the past 20-50 years is 0.7, and the bridge built in the past 50 years is 1. The higher the design standard, the stronger the crack resistance of the bridge and the lower the risk of cracks. The bridge maintenance factor reflects the maintenance of the bridge. The comprehensive score of bridge cracks combines these factors and corresponding weight coefficients, where each factor is assigned a different weight according to its importance in the crack generation and expansion process, and the weight can be adjusted according to actual conditions. Based on the calculated comprehensive score of bridge cracks, different length thresholds can be set. If the comprehensive score is low, it means that the bridge will not have many long cracks. If the comprehensive score is high, it means that the bridge will have more long cracks.

[0093] In some embodiments of the present application, when determining the length threshold according to the comprehensive score of the bridge crack, and dividing the length data into the first length data and the second length data according to the set length threshold, it includes: when When it is greater than or equal to 0.8, it is set as the first length threshold. When it is less than 0.8, it is set as the second length threshold, and the data of the target crack length data greater than or equal to the length threshold is divided into the first length data, and the data of the target crack length data less than the length threshold is divided into the second length data.

[0094] It is understandable that, for example, a concrete bridge built in the past 20-50 years has a recommended maintenance frequency of 6 times and an actual maintenance frequency of 3 times, has been in use for 30 years, and has a service life of 50 years. and is 0.3, and is 0.2, then the comprehensive score of bridge cracks is equal to 0.66; a concrete bridge was built in the past 50 years, the recommended maintenance times are 10 times, the actual maintenance times are 2 times, it has been used for 45 years, and the service life is 50 years. and is 0.3, and is 0.2, then the comprehensive score of bridge cracks is equal to 0.88. When the comprehensive score of bridge cracks is greater than or equal to 0.8, it is judged that the bridge will have more long cracks. At this time, the first length threshold is appropriately adjusted according to the length of the cracks. When the comprehensive score of bridge cracks is less than 0.8, it is judged that the bridge will have fewer long cracks. At this time, the second length threshold is appropriately adjusted according to the length of the cracks. The length threshold is set according to the comprehensive score of bridge cracks, and the target crack length data is rationally separated into the first length data and the second length data. The first length data is represented as the data of long cracks, and the second length data is represented as the data of short cracks. The first length data and the second length data are respectively constructed as the first crack set and the second crack set through the obtained first length data and the second length data, thereby further reducing the error and subjectivity of the measurement process.

[0095] In some embodiments of the present application, when the first set of cracks and the second set of cracks are aggregated using a clustering algorithm to obtain the first cluster set and the second cluster set respectively, the method includes: setting the number of clusters k=2, randomly selecting two first length data in the first set of cracks as the first cluster center and the second cluster center, respectively calculating the distance to the first cluster center and the second cluster center for each first length data in the first set of cracks, and assigning them to the cluster with the closest distance. When the distances are the same, they are assigned to the cluster where the first cluster center is located. The distance to the first cluster center and the second cluster center for each first length data in the first set of cracks is respectively calculated, and is obtained by the following formula:

[0096] ;

[0097] ;

[0098] in, Represents any first length data in the first set of cracks, represents the center of the first cluster, represents the second cluster center, Represents the distance from any first-length data in the first set to the center of the first cluster, It represents the distance from any first length data in the first set to the center of the second cluster. According to the allocation result, the first cluster set and the second cluster set are obtained.

[0099] It can be understood that, for example: the first set of cracks is [7,9,11,14,6,16], the number of clusters is set to 2, 9 and 14 are randomly selected as the first cluster center and the second cluster center, the distance from each crack length data to the first cluster center and the second cluster center is calculated, and a 7 is selected as an example, the distance from the first cluster center is 2, and the distance from the second cluster center is 7, then 7 is assigned to the cluster where the first cluster center is located, and finally all data are calculated to obtain the first cluster set, and the first cluster set is divided into two clusters. The method of obtaining the second cluster set is the same as the method of the first cluster set, which will not be repeated here. The aggregation algorithm can effectively aggregate the first set of cracks and the second set of cracks, respectively, reducing the error and contingency of the measurement.

[0100] In some embodiments of the present application, when the first data and the second data are obtained according to the first cluster set and the second cluster set respectively, and the crack length factor is calculated according to the first data and the second data, it includes: taking the average value of the cluster where the center of the first cluster in the first cluster set is located to obtain a first mean value, taking the average value of the cluster where the center of the second cluster in the first cluster set is located to obtain a second mean value, obtaining a third mean value and a fourth mean value according to the second cluster set, and calculating the crack length factor according to the first mean value, the second mean value, the third mean value and the fourth mean value, and the crack length factor is obtained by the following formula

[0101]

[0102] in, represents the crack length factor, represents the first mean, represents the second mean, represents the third mean, represents the fourth mean, , , and is the weight coefficient, and .

[0103] It can be understood that taking the average of the cluster where the center of the first cluster in the first cluster set is located to obtain the first mean, and taking the average of the cluster where the center of the second cluster in the first cluster set is located to obtain the second mean, can reduce the complexity of processing the first length data and the second length data without losing the first length data and the second length data. Taking the average value within the cluster can reduce the interference of accidental data on the results, and improve the accuracy of the crack length factor. The third mean and the fourth mean are obtained in the same way as the first mean and the second mean, respectively, and will not be repeated here. The long cracks and short cracks are rationally merged, and the crack length factor calculated by the first mean, the second mean, the third mean and the fourth mean reflects the specific value of the bridge crack length and improves the accuracy of the data.

[0104] In some embodiments of the present application, when comparing the crack length factor with the historical crack length factor of bridge cracks, and judging whether to correct the crack length factor and obtain the target crack length factor based on the comparison result, it includes: when the crack length factor is greater than or equal to the minimum value of the historical crack length factor, judging not to correct the crack length factor, and determining the target crack length factor based on the crack length factor; when the crack length factor is less than the minimum value of the historical crack length factor, judging to correct the crack length factor; when it is judged that the crack length factor is corrected, comparing the crack length factor with the mean of the historical crack length factors, and determining the final correction factor based on the factor similarity.

[0105] It can be understood that by comparing the crack length factor with the historical crack length factor, the consistency of the processed data can be ensured. If the currently measured crack length factor deviates from the reasonable range of the historical crack length factor, it needs to be corrected to ensure that the crack length factor conforms to the actual situation, effectively reduce the impact of abnormal data caused by measurement factors, improve the accuracy of crack length measurement, avoid subjective bias caused by human intervention, and compare the crack length factor with the mean of historical crack length factors to avoid accidental errors in the data and enhance the objectivity and fairness of data analysis.

[0106] In some embodiments of the present application, when comparing the crack length factor with the average value of the historical crack length factors and determining the final correction factor according to the factor similarity, it includes:

[0107] The factor similarity is given by the following formula:

[0108]

[0109]

[0110] in, represents the standard deviation of the historical crack length factor, represents the number of historical crack length factors, represents the i-th historical crack length factor, is the mean value of the historical crack length factor, represents the crack length factor, Represents factor similarity, when When it is greater than or equal to 0.7, the first preset correction factor is used as the final correction factor of the crack length factor. When it is less than 0.7 and greater than or equal to 0, the second preset correction factor is used as the final correction factor of the crack length factor. When it is less than 0, the third preset correction factor is used as the final correction factor of the crack length factor, and the target crack length factor is determined according to the product value of the final correction factor and the crack length factor.

[0111] It is understandable that, for example: the historical crack length factors are [6, 8, 10, 12, 14], and the resulting crack length factor is 5, then it is judged that it needs to be corrected. The mean of the historical crack length factors is 10, and the standard deviation is 2, then the similarity is -2. The first preset correction factor is preferably 1.3, the second preset correction factor is preferably 1.1, and the third preset correction factor is preferably 1.5, which can be adaptively adjusted according to actual needs. The correction process can automatically adapt to the distribution of historical crack length factors and the difference in current crack length factors. With the accumulation of historical crack length factors, the correction process can be continuously adjusted and optimized, thereby improving the accurate target crack length factor.

[0112] In some embodiments of the present application, when establishing a length data correction factor for a corresponding data detection area based on the first length data and the second length data, it includes: determining the first standard crack length data and the second standard crack length data corresponding to each data detection area, comparing the first length data and the second length data with the first standard crack length data and the second standard crack length data, respectively, when the first length data and the second length data are respectively less than or equal to the first standard crack length data and the second standard crack length data, determining that the corresponding data detection area generates a first correction factor, when all first length data are less than or equal to the first standard crack length data and all second length data are less than or equal to the second standard crack length data, determining that the corresponding data detection area generates a first correction factor, when all first length data are greater than the first standard crack length data and all second length data are greater than the second standard crack length data, determining that the corresponding data detection area generates a second correction factor, and when there are one or more first length data that are not equal to the first standard crack length data and there are one or more second length data that are not equal to the second standard crack length data, determining that the corresponding data detection area generates a third correction factor.

[0113] It can be understood that the first standard crack length data and the second standard crack length data are the length standards for long cracks and short cracks in bridge cracks. The first standard crack length data and the second standard crack length data are derived from the highway bridge technical condition assessment standard JTG / T H21-2017. By determining the number of data detection areas that meet the first standard crack length data and the second standard crack length data, the accuracy of the obtained first length data and the second length data is improved, further ensuring the accuracy of the calculated final crack length data.

[0114] In some embodiments of the present application, when setting the correction constant of the target crack length factor according to the first correction number, the second correction number and the third correction number, it includes: determining the first correction number, the second correction number and the third correction number for normalization processing to obtain a normalized value J, the preset correction constants are the first preset correction constant, the second preset correction constant and the third preset correction constant, when J≤2, the first preset correction constant is used as the correction constant of the target crack length factor, when 2<J≤3.5, the second preset correction constant is used as the correction constant of the target crack length factor, when 3.5<J, the third preset correction constant is used as the correction constant of the target crack length factor, and the final crack length data is determined according to the product value of the correction constant and the target crack length.

[0115] It can be understood that the normalization processing includes obtaining the difference between the maximum value and the minimum value of the first corrected number, the second corrected number and the third corrected number, dividing the result of subtracting the minimum value from the first corrected number by the difference to obtain a first normalized value, the second normalized value is the result of subtracting the minimum value from the second corrected number divided by the difference, and the third normalized value is the result of subtracting the minimum value from the third corrected number divided by the difference, taking the average of the first normalized value, the second normalized value and the third normalized value to obtain a normalized value J, and selecting the corresponding preset correction constant based on J, thereby realizing dynamic optimization of the target crack length, ensuring the accuracy of the final crack length data, and avoiding inaccurate final crack length data caused by measurement errors and randomness of measurement results.

[0116] In summary, the beneficial effects of the present invention are as follows: compared with the prior art, the beneficial effects of the present invention are as follows: by preprocessing and optimizing the initial crack length data, the error in the measurement process can be effectively removed, the accuracy of the target crack length data obtained can be ensured, and the influence of accidental errors and human factors on the accuracy of the crack length data can be reduced, thereby improving the accuracy of crack detection. By clustering the target crack length data and distinguishing the crack data of different lengths, redundant data can be effectively eliminated, the availability and reliability of the data can be improved, and the crack length factor can be compared with the historical crack length factor to find abnormal fluctuations in the data, and on this basis, the crack length factor can be corrected. The accuracy of the crack detection results can be further improved, and the result errors caused by single detection errors can be avoided. By establishing a length data correction factor, the influence of data deviation on the final crack length data can be avoided, the refinement of the correction process can be improved, the length data of the bridge cracks can be accurately reflected, and the risk of bridge accidents can be reduced.

[0117] In another preferred embodiment based on the above embodiment, refer to Figure 2 As shown, this embodiment provides a data processing system for bridge crack detection, which is used to apply the above-mentioned data processing method for bridge crack detection, including:

[0118] Acquisition unit: deploy multiple data detection areas for the bridge cracks to be detected, obtain the initial crack length data corresponding to each data detection area, pre-process the initial crack length data, and determine the target crack length data according to the pre-processing results;

[0119] Aggregation unit: divides the target crack length data into first length data and second length data according to a set length threshold, extracts the same type of first length data and second length data from each data detection area, performs optimization processing on them respectively, and constructs a first crack set and a second crack set;

[0120] The first set of cracks and the second set of cracks are aggregated by using a clustering algorithm to obtain a first cluster set and a second cluster set respectively, first data and second data are obtained according to the first cluster set and the second cluster set respectively, and a crack length factor is calculated according to the first data and the second data;

[0121] The first correction unit is used to compare the crack length factor with the historical crack length factor of the bridge crack, and determine whether to correct the crack length factor according to the comparison result and obtain the target crack length factor;

[0122] A second correction unit: establishing a length data correction factor for the corresponding data detection area according to the first length data and the second length data, wherein the length data correction factor includes a first correction factor, a second correction factor and a third correction factor;

[0123] A length data confirmation unit: counting a first correction number of a first correction factor, counting a second correction number of a second correction factor, counting a third correction number of a third correction factor, and setting a correction constant of a target crack length factor according to the first correction number, the second correction number, and the third correction number;

[0124] The target crack length factor is corrected according to the correction constant to obtain the final crack length data.

[0125] It can be understood that the system improves the accuracy and reliability of the final crack length data by performing multi-stage processing on the detection data of bridge cracks. The acquisition unit deploys multiple data detection areas, collects the initial crack length data of each area and performs preprocessing to determine the target crack length data. The aggregation unit divides the target crack length data into two categories according to the set length threshold: the first length data and the second length data, and constructs the first set of cracks and the second set of cracks. The clustering algorithm is applied to the first set of cracks and the second set of cracks to obtain the first cluster set and the second cluster set respectively, and then the crack length factor is calculated based on the clustering results. The first correction unit compares the crack length factor with the historical crack length factor to determine whether correction is needed to obtain the target crack length factor. The second correction unit establishes correction factors for different data detection areas through the first and second length data, and sets the correction constant according to the statistical results to perform the final correction on the target crack length factor to ensure that accurate final crack length data is obtained, which helps to improve the reliability and accuracy of the bridge crack data results.

[0126] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0127] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0128] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0129] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A data processing method for bridge crack detection, characterized in that: include: Deploy multiple data detection areas for the bridge cracks to be detected, obtain initial crack length data corresponding to each data detection area, pre-process the initial crack length data, and determine target crack length data according to the pre-processing results; The target crack length data is divided into first length data and second length data according to a set length threshold, and the first length data and second length data of the same type are extracted from each data detection area, and are respectively optimized to construct a first crack set and a second crack set; The first set of cracks and the second set of cracks are aggregated by using a clustering algorithm to obtain a first cluster set and a second cluster set respectively, first data and second data are obtained according to the first cluster set and the second cluster set respectively, and a crack length factor is calculated according to the first data and the second data; The number of clusters k is set to 2, two first length data in the first set of cracks are randomly selected as the first cluster center and the second cluster center, the first mean is obtained by taking the average value of the cluster where the first cluster center is located in the first cluster set, the second mean is obtained by taking the average value of the cluster where the second cluster center is located in the first cluster set, the first mean and the second mean are used as the first data, a third mean and a fourth mean are obtained according to the second cluster set, and the third mean and the fourth mean are used as the second data, wherein the third mean and the fourth mean are obtained in the same manner as the first mean and the second mean, respectively; Comparing the crack length factor with the historical crack length factor of the bridge crack, and judging whether to correct the crack length factor and obtain a target crack length factor according to the comparison result; Establishing a length data correction factor for the corresponding data detection area according to the first length data and the second length data, wherein the length data correction factor includes a first correction factor, a second correction factor and a third correction factor; Counting a first correction quantity of the first correction factor, counting a second correction quantity of the second correction factor, counting a third correction quantity of the third correction factor, and setting a correction constant of the target crack length factor according to the first correction quantity, the second correction quantity, and the third correction quantity; The target crack length factor is corrected according to the correction constant to obtain final crack length data.

2. The data processing method for bridge crack detection according to claim 1, characterized in that: When the target crack length data is divided into first length data and second length data according to a set length threshold, it includes: The comprehensive score of bridge cracks is calculated based on the bridge service factor, bridge material factor, design standard factor and bridge maintenance factor. The comprehensive score of bridge cracks is calculated by the following formula: in, represents the bridge service factor, Indicates the current year, Indicates the year the bridge was built. represents the design service life of the bridge, represents the bridge maintenance factor, Indicates the actual number of maintenance times. Indicates the recommended maintenance times for the bridge. represents the bridge material factor, represents the design standard factor, , , and is the weight coefficient, and , It represents the comprehensive score of bridge cracks; The length threshold is determined according to the comprehensive score of the bridge cracks, and is divided into first length data and second length data according to the set length threshold.

3. The data processing method for bridge crack detection according to claim 2 is characterized in that: When the length threshold is determined according to the comprehensive score of the bridge crack, and the length threshold is divided into first length data and second length data according to the set length threshold, it includes: when When it is greater than or equal to 0.8, it is set as the first length threshold; when When it is less than 0.8, it is set as the second length threshold; Classify the target crack length data that is greater than or equal to the length threshold as the first length data; The target crack length data that is smaller than the length threshold is classified as the second length data.

4. The data processing method for bridge crack detection according to claim 3 is characterized in that: When the first set of fractures and the second set of fractures are aggregated by using a clustering algorithm to obtain a first cluster set and a second cluster set respectively, the method includes: For each first length data in the first set of cracks, respectively, the distance to the center of the first cluster and the center of the second cluster is calculated, and the first length data is assigned to the cluster with the closest distance; when the distances are the same, the first length data is assigned to the cluster where the center of the first cluster is located; The distances of each first length data in the first set of cracks to the center of the first cluster and the center of the second cluster are calculated respectively, and are obtained by the following formula: in, represents any first length data in the first set of cracks, represents the first cluster center, represents the second cluster center, represents the distance from any first-length data in the first set to the center of the first cluster, Indicates the distance from any first-length data in the first set to the center of the second cluster; According to the allocation result, the first cluster set and the second cluster set are obtained.

5. The data processing method for bridge crack detection according to claim 4 is characterized in that: When first data and second data are obtained respectively according to the first cluster set and the second cluster set, and a crack length factor is calculated according to the first data and the second data, the method includes: The crack length factor is calculated based on the first mean, the second mean, the third mean and the fourth mean. The crack length factor is obtained by the following formula: in, represents the crack length factor, represents the first mean, represents the second mean, represents the third mean, represents the fourth mean, , , and is the weight coefficient, and .

6. The data processing method for bridge crack detection according to claim 5, characterized in that: When comparing the crack length factor with the historical crack length factor of the bridge crack, and judging whether to correct the crack length factor and obtain the target crack length factor according to the comparison result, the method includes: When the crack length factor is greater than or equal to the minimum value of the historical crack length factors, it is determined that the crack length factor is not to be corrected, and the crack length factor is determined as a target crack length factor; When the crack length factor is less than the minimum value of the historical crack length factor, determining to correct the crack length factor; When it is determined to correct the crack length factor, the crack length factor is compared with the average of the historical crack length factors, and a final correction factor is determined according to the factor similarity.

7. The data processing method for bridge crack detection according to claim 6, characterized in that: When comparing the crack length factor with the average value of the historical crack length factor and determining the final correction factor according to the factor similarity, it includes: The factor similarity is obtained by the following formula: in, represents the standard deviation of the historical crack length factor, represents the number of historical crack length factors, represents the i-th historical crack length factor, is the mean value of the historical crack length factor, represents the crack length factor, Indicates factor similarity; when When it is greater than or equal to 0.7, the first preset correction factor is used as the final correction factor of the crack length factor; when When it is less than 0.7 and greater than or equal to 0, the second preset correction factor is used as the final correction factor of the crack length factor; when When it is less than 0, the third preset correction factor is used as the final correction factor of the crack length factor; The target crack length factor is determined according to the product value of the final correction factor and the crack length factor.

8. The data processing method for bridge crack detection according to claim 7, characterized in that: When establishing a length data correction factor for a corresponding data detection area according to the first length data and the second length data, it includes: Determine the first standard crack length data and the second standard crack length data corresponding to each data detection area; comparing the first length data and the second length data with the first standard crack length data and the second standard crack length data respectively; When all the first length data are less than or equal to the first standard crack length data and all the second length data are less than or equal to the second standard crack length data, determining a corresponding data detection area to generate the first correction factor; When all the first length data are greater than the first standard crack length data and all the second length data are greater than the second standard crack length data, determining the corresponding data detection area to generate the second correction factor; When there are one or more first length data that are not equal to the first standard crack length data and there are one or more second length data that are not equal to the second standard crack length data, the corresponding data detection area is determined to generate the third correction factor.

9. The data processing method for bridge crack detection according to claim 8, characterized in that: When the correction constant of the target crack length factor is set according to the first correction amount, the second correction amount and the third correction amount, it includes: Determine a first correction number, a second correction number, and a third correction number, perform a normalization process on them, and obtain a normalized value J; The preset calibration constants are a first preset calibration constant, a second preset calibration constant and a third preset calibration constant; When J≤2, the first preset correction constant is used as the correction constant of the target crack length factor; When 2<J≤3.5, the second preset correction constant is used as the correction constant of the target crack length factor; When 3.5<J, the third preset correction constant is used as the correction constant of the target crack length factor; The final crack length data is determined according to the product value of the correction constant and the target crack length factor.

10. A data processing system for bridge crack detection, used for applying the data processing method for bridge crack detection according to any one of claims 1 to 9, characterized in that: include: Acquisition unit: deploy multiple data detection areas for the bridge cracks to be detected, obtain initial crack length data corresponding to each data detection area, pre-process the initial crack length data, and determine target crack length data according to the pre-processing results; Aggregation unit: dividing the target crack length data into first length data and second length data according to a set length threshold, extracting the same type of first length data and second length data from each data detection area, respectively performing optimization processing and constructing a first crack set and a second crack set; The first set of cracks and the second set of cracks are aggregated by using a clustering algorithm to obtain a first cluster set and a second cluster set respectively, first data and second data are obtained according to the first cluster set and the second cluster set respectively, and a crack length factor is calculated according to the first data and the second data; The number of clusters k is set to 2, two first length data in the first set of cracks are randomly selected as the first cluster center and the second cluster center, the first mean is obtained by taking the average value of the cluster where the first cluster center is located in the first cluster set, the second mean is obtained by taking the average value of the cluster where the second cluster center is located in the first cluster set, the first mean and the second mean are used as the first data, a third mean and a fourth mean are obtained according to the second cluster set, and the third mean and the fourth mean are used as the second data, wherein the third mean and the fourth mean are obtained in the same manner as the first mean and the second mean, respectively; A first correction unit: compares the crack length factor with the historical crack length factor of the bridge crack, determines whether to correct the crack length factor according to the comparison result, and obtains a target crack length factor; A second correction unit: establishing a length data correction factor for a corresponding data detection area according to the first length data and the second length data, wherein the length data correction factor includes a first correction factor, a second correction factor and a third correction factor; A length data confirmation unit: counting a first correction number of the first correction factor, counting a second correction number of the second correction factor, counting a third correction number of the third correction factor, and setting a correction constant of the target crack length factor according to the first correction number, the second correction number, and the third correction number; The target crack length factor is corrected according to the correction constant to obtain final crack length data.

Images