A crack detection method and system for highway construction

Abstract

The present invention belongs to the technical field of construction crack detection, and specifically discloses a crack detection method and system for highway construction. The present invention uses a multispectral thermal imaging sensor to synchronously collect multi-band data and utilizes multi-source registration and fusion technology to integrate surface morphology, material reflection, and deep thermal radiation characteristics, thereby enhancing the crack identification capability in complex environments. Furthermore, through dynamic correlation analysis of crack edge continuity and thermal radiation fluctuation characteristics, transient interference is accurately removed, solving the problem of high false alarms. At the same time, a scoring model is constructed by combining geometric and thermodynamic parameters to quantify the degree of damage and expansion trend. Finally, the dynamic change trend of thermodynamic parameters is used to predict the risk of crack expansion, triggering a graded warning signal to form a closed-loop chain of perception-anti-interference-assessment-decision-making, realizing full-process optimization from data fusion and interference suppression to risk assessment and maintenance response, thereby ensuring the accuracy of highway crack detection.

Description

Technical Field

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

[0002] Cracks are a common early-stage defect on highways and can be caused by construction defects, aging materials, excessive loads, or environmental factors. If not detected and addressed promptly, cracks can expand and lead to subgrade damage and pavement collapse, threatening driving safety.

[0003] Existing technologies, such as the Chinese invention patent application with application number CN202110551750.8, disclose a roadbed crack detection, identification and repair method based on infrared thermal imaging analysis. The method fixes the infrared thermal imaging equipment and visible light imaging equipment by laying guide rails, synchronously collects infrared thermal images and visible light images of the road surface, manually processes the infrared data to generate temperature contour maps, combines visible light images for comparison to identify hidden cracks, and predicts the development trend of cracks based on temperature distribution.

[0004] Another example of existing technology is a method for detecting the degree of crack development in asphalt pavement based on infrared thermal image analysis, which is disclosed in the Chinese invention patent application with application number CN201680089245.3. The method establishes a crack development degree model, uses infrared thermal images to obtain the measured temperature difference data between the crack area and the pavement, combines the atmospheric temperature correction model to estimate the degree of crack development, and simultaneously integrates grayscale information and temperature information for detection.

[0005] For asphalt pavement, the core of the Chinese invention patent application with application number CN202110551750.8 is the dual-modal data fusion of infrared and visible light and manual assisted analysis, which is suitable for small-scale static detection, but relies on guide rails to limit flexibility. The core of the Chinese invention patent application with application number CN201680089245.3 is the static temperature difference model and atmospheric temperature compensation, but relies on a preset model. Both of these technical solutions are based on the fusion of infrared thermal imaging and visible light images, and identify cracks through temperature difference or temperature distribution. There are still the following deficiencies: 1. The existing solution only fuses infrared thermal imaging and visible light images, and does not fully utilize multi-spectral or thermodynamic dynamic parameters. The data fusion dimension is still insufficient.

[0006] 2. Relying on static temperature thresholds or manual corrections, it cannot respond to dynamic interference such as steam and dust in real time. In scenarios where steam and oil stains mix after heavy rain, the false alarm rate is high.

[0007] 3. Existing technologies only qualitatively assess cracks through temperature differences or grayscale information, lack of risk quantification, and are not convenient for subsequent processing. Summary of the Invention

[0008] In view of this, a crack detection method and system for highway construction are proposed to address the limitations of existing technologies in preventing false alarms due to water vapor, such as poor environmental adaptability, sensitivity to optical contamination, and rough feature analysis.

[0009] The purpose of the present invention can be achieved through the following technical solutions: The first aspect of the present invention provides a crack detection method for highway construction, which includes the following steps: S1, synchronously collecting multi-band image data of the asphalt layer surface through a multispectral thermal imaging sensor, and performing alignment and fusion, wherein the multi-band includes a visible light band, a near-infrared band and a long-wave infrared band.

[0010] S2. Based on the correlation analysis of edge continuity characteristics and thermal radiation fluctuation characteristics, false crack interference data is eliminated and effective multi-band image data is obtained.

[0011] S3. Extract the geometric parameters and thermodynamic parameters of the crack from the effective multi-band image data, and calculate the current comprehensive risk factor through a weighted scoring model.

[0012] S4. Generate a graded thermal map based on the spatial distribution of the current comprehensive risk factor, and trigger a graded warning signal in combination with the expansion trend to generate a crack detection report.

[0013] The second aspect of the present invention provides a crack detection system for highway construction, which includes the following modules: a multispectral thermal imaging acquisition module, which synchronously collects multi-band image data of the asphalt layer surface through a multispectral thermal imaging sensor and performs alignment and fusion.

[0014] The crack interference identification and elimination module eliminates false crack interference data based on the correlation analysis of edge continuity features and thermal radiation fluctuation features, and obtains effective multi-band image data.

[0015] The crack risk analysis module extracts the geometric parameters and thermodynamic parameters of the cracks from the effective multi-band image data and calculates the current comprehensive risk factor through a weighted scoring model.

[0016] The crack report generation terminal generates a graded thermal map based on the current spatial distribution of the comprehensive hazard factor, and triggers a graded warning signal based on the expansion trend to generate a crack detection report.

[0017] Compared with the existing technology, the beneficial effects of the present invention are as follows: (1) The present invention uses a multi-spectral thermal imaging sensor to synchronously collect multi-band data, and uses multi-source registration and fusion technology to integrate surface morphology, material reflection and deep thermal radiation characteristics, thereby enhancing the ability to identify cracks in complex environments. By dynamically correlating and analyzing the continuity of crack edges and the fluctuation characteristics of thermal radiation, transient interference is accurately removed, solving the problem of high false alarms. A scoring model is constructed by combining geometric and thermodynamic parameters to quantify the degree of damage and expansion trend, generate a thermal map, and predict risk triggering warnings, thereby achieving full process optimization and improving detection accuracy and environmental adaptability.

[0018] (2) The present invention overcomes the limitations of a single sensor in complex environments by synchronously collecting multiple bands such as visible light, near infrared, and long-wave infrared, significantly improving the contrast between cracks and background, and effectively solving the problems of missed detection and false detection caused by insufficient data dimensions in traditional methods, providing a high-precision, multi-dimensional data foundation for subsequent dynamic interference elimination and quantitative evaluation.

[0019] (3) The present invention removes dynamic interference by analyzing the correlation between edge continuity characteristics and thermal radiation fluctuation characteristics, breaking through the limitations of traditional static thresholds or manual corrections, and achieving real-time identification and removal of interference, making it easier to accurately distinguish between real cracks and transient interference such as steam and dust, reducing false alarm rates and ensuring the reliability of subsequent crack detection.

[0020] (4) The present invention calculates the current comprehensive risk factor by building a weighted scoring model based on the geometric morphology of the cracks and the thermodynamic dynamic parameters, thereby achieving a quantitative risk assessment of the cracks and making up for the defect of relying solely on qualitative assessment of temperature difference. At the same time, by generating a spatial distribution thermal map and a crack detection report, the risk level distribution is intuitively quantified, significantly reducing the engineering safety hazards caused by assessment deviations. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 The figure is a schematic flow chart of the steps for implementing the method of the present invention.

[0023] Figure 2 This is a schematic diagram of the system module structure connection of the present invention.

[0024] Figure 3 Schematic diagram of the process of eliminating false crack interference data in the present invention. DETAILED DESCRIPTION

[0025] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0026] See also Figure 1 As shown, the present invention provides a crack detection method for highway construction, which includes: S1, synchronously collecting multi-band image data of the asphalt layer surface through a multispectral thermal imaging sensor, and performing alignment and fusion, wherein the multi-band includes a visible light band, a near-infrared band, and a long-wave infrared band.

[0027] Specifically, the specific execution process of the step S1 includes: S11, synchronously collecting the surface image of the asphalt layer in the visible light band, near infrared band and long-wave infrared band through a multispectral sensor.

[0028] S12. Use an image registration algorithm to spatially align and fuse the multi-band images to generate a fused image containing temperature gradient distribution and texture features.

[0029] S13. Mark the geographic location coordinates of each acquisition area and bind the fused image with the coordinate information.

[0030] It should be added that the registration fusion is performed using an image registration algorithm, which is a relatively mature algorithm available, and its specific execution process will not be described here.

[0031] The embodiments of the present invention overcome the limitations of a single sensor in complex environments by performing synchronous acquisition of multiple bands such as visible light, near-infrared, and long-wave infrared, significantly improving the contrast between cracks and background, and effectively solving the problems of missed detection and false detection caused by insufficient data dimensions in traditional methods, providing a high-precision, multi-dimensional data foundation for subsequent dynamic interference elimination and quantitative evaluation.

[0032] See also Figure 3 As shown in S2, the correlation analysis based on edge continuity characteristics and thermal radiation fluctuation characteristics is used to eliminate false crack interference data and obtain effective multi-band image data.

[0033] Specifically, the specific elimination process of removing false crack interference data includes: S21, grayscale processing of the visible light band image, setting the sliding window size based on the minimum apparent width of the crack, traversing and calculating the texture roughness of each pixel neighborhood, and marking the area with texture roughness lower than the preset threshold range as a low roughness area.

[0034] S22. In the thermal radiation image sequence synchronously collected in the near-infrared band, extract the time-series fluctuation data of the thermal radiation intensity of the low-roughness area, calculate its standard deviation and mean offset, and at the same time calculate the deviation value between the temperature distribution of the shadow area in the long-wave infrared band image and the corresponding brightness distribution of the visible light image.

[0035] S23. Extract the edge continuity features of the detection area from the visible light band image, combine it with the thermal radiation intensity fluctuation characteristics of the near-infrared band, construct the texture-thermal radiation correlation matrix, and calculate the correlation coefficient between the two.

[0036] S24. Perform interference determination on the standard deviation, mean offset, deviation value and correlation using pre-set interference determination rules, and synchronously delete image data in the visible light, near infrared and long-wave infrared bands for areas determined to be interference areas.

[0037] Furthermore, the size of the sliding window is associated with the minimum apparent width of the target crack, specifically satisfying: , is the sliding window width, represents the minimum apparent width of the target crack, and the traversal step size of the sliding window is set to 1 / 2 of the window width to ensure that there is a 50% overlap area between adjacent windows.

[0038] It should be added that the preset threshold range of texture roughness can be determined by the roughness distribution statistics of historical crack samples.

[0039] Furthermore, the mean shift is the absolute deviation percentage of the average value of the thermal radiation intensity time series data within the observation period relative to the thermal radiation intensity value of the surrounding area.

[0040] It's important to note that real cracks can cause significant variations in thermal radiation due to different materials or structural issues, while artificial cracks, such as stains or shadows, may have smaller fluctuations or distinct patterns. However, a threshold must be set; too high or too low can be problematic. The same real crack can have different thermal properties, resulting in significant mean shift.

[0041] Understandably, due to the irregularities of the internal structure and the complexity of the heat conduction pathways, real cracks typically experience high standard deviations in their thermal radiation fluctuations. However, due to the homogeneity of surface deposits such as paint and dust, thermal radiation fluctuations are more gradual, with significantly lower standard deviations than real cracks. Furthermore, differences in thermal inertia between surface deposits and the substrate can lead to significant mean shifts in thermal radiation time series data, such as differences in the rates at which deposits absorb and dissipate heat. Therefore, analysis is conducted using two parameters: standard deviation and mean shift.

[0042] Furthermore, the deviation between temperature distribution and brightness distribution is expressed by the formula Calculate, where Indicates the deviation between temperature distribution and brightness distribution, is the temperature standard deviation of the shadow area in the long-wave infrared image, is the brightness standard deviation of the corresponding area of ​​the visible light image, It is the band response coefficient determined by calibration experiment, and its unit is ℃ / grayscale value. Its physical meaning is the temperature standard deviation corresponding to the unit brightness standard deviation.

[0043] The calibration experiment refers to collecting multiple sets of shadow-free samples under a standard lighting environment, calculating their temperature and brightness standard deviations, and fitting them using the least squares method. value, so that and The units are consistent.

[0044] Furthermore, the correlation coefficient can be calculated by normalizing the edge continuity feature vector and the thermal radiation fluctuation data sequence using the Pearson correlation coefficient formula. The specific calculation formula is: , represents the correlation coefficient, is the edge continuity feature, is the thermal radiation fluctuation characteristic, is the covariance operation, and They represent the standard deviation of edge continuity characteristics and thermal radiation fluctuation characteristics respectively.

[0045] Understandably, the Pearson correlation coefficient normalizes the covariance by dividing it by the product of the standard deviations, eliminating dimensionality and making the results more comparable. It also ensures that changes in the position and scale of the variables do not affect the coefficient. It accurately reflects the strength and direction of the linear relationship between edge continuity characteristics and thermal radiation fluctuation characteristics, providing a reliable quantitative basis for analyzing the relationship between the two and facilitating intuitive judgment of the strength and direction of the correlation.

[0046] Furthermore, the specific interference determination process includes: if the standard deviation of the thermal radiation intensity is lower than the lower limit of the standard deviation reference range of the crack characteristic model and its mean offset exceeds a preset offset threshold, it is determined that there is surface attachment interference.

[0047] If the deviation value exceeds a preset adjustment threshold, it is determined that optical shadow artifact interference exists.

[0048] If the correlation coefficient is lower than a preset correlation coefficient threshold, it is determined that there is stain or foreign matter interference.

[0049] In a specific embodiment, the thermal radiation standard deviation reference range, preset correlation coefficient threshold and preset adjustment threshold are generated by joint calibration of multimodal crack data and dynamically updated in the crack characteristic model. Exemplarily, the thermal radiation standard deviation reference range is determined by the following steps: (a) collecting thermal radiation time series fluctuation data of a known real crack area.

[0050] (b) Calculate the standard deviation of each true crack area and generate a standard deviation distribution histogram.

[0051] (c) The 5th and 95th percentiles of the distribution histogram are used as the lower and upper limits of the reference range, respectively, to form the standard deviation reference range.

[0052] It's important to explain that by using the 5th to 95th percentile range, we can exclude extreme standard deviations, both high and low, and retain a data range consistent with the fluctuation characteristics of actual crack thermal radiation. Low standard deviations below the 5th percentile may correspond to surface deposits such as paint and stains, while high standard deviations above the 95th percentile may be caused by non-crack heat sources, such as localized thermal radiation anomalies. Both are excluded. Furthermore, this percentile range can be dynamically adjusted based on updates to the actual crack sample library.

[0053] The embodiment of the present invention performs dynamic interference elimination by conducting correlation analysis between edge continuity features and thermal radiation fluctuation features, breaking through the limitations of traditional static thresholds or manual corrections, and realizing real-time identification and elimination of interference, facilitating the accurate distinction between real cracks and transient interferences such as steam and dust, reducing false alarm rates and ensuring the reliability of subsequent crack detection.

[0054] S3. Extract the geometric parameters and thermodynamic parameters of the crack from the effective multi-band image data, and calculate the current comprehensive risk factor through a weighted scoring model.

[0055] Specifically, the geometric parameters of the crack include crack length, average crack width, horizontal crack development rate, crack branching density and crack width gradient with depth. The specific extraction process includes: A1, using the Canny edge detection algorithm to identify the crack center in the visible light band image, measuring the longest continuous pixel distance along the crack center line, and converting it into actual length in combination with the image resolution as the crack length.

[0056] A2. Take multiple sampling points perpendicular to the crack centerline and calculate the average value as the average crack width.

[0057] A3. Measure the width of the crack in sections along its development direction, and use linear regression to fit the slope of the width versus extension distance, which is used as the horizontal development rate of the crack width.

[0058] A4. Count the number of crack bifurcation points from the visible light band image and compare it with the area of ​​the crack region to obtain the crack branch density.

[0059] A5. Extract the corresponding crack cross-sectional profile from the visible light band image, then measure the width layer by layer along the depth direction and calculate its gradient with depth.

[0060] It is understandable that the Canny edge detection algorithm is a relatively mature algorithm currently available, and its specific detection process and recognition principle will not be described in detail here.

[0061] It should be added that the layer spacing along the depth direction is dynamically adjusted according to the apparent width of the crack, and the change gradient is quantified by the ratio of the width difference between adjacent layers to the depth increment.

[0062] Cross sections are taken at equal intervals along the extension direction of the crack centerline, and the intervals are no more than twice the apparent width of the crack.

[0063] For example, dynamic interval stratification can be divided according to the following rules: surface interval, depth 0-5mm, stratification interval 0.5mm; middle interval, depth 5-20mm, stratification interval 1mm; deep interval, depth >20mm, stratification interval 2mm.

[0064] Furthermore, the specific calculation process of the change gradient includes: B1, cutting cross sections at equal intervals along the extension direction of the crack centerline, and layering at dynamic intervals along the depth direction on each cross section.

[0065] B2. Measure the maximum and minimum width values ​​within each layer and calculate the intra-layer width fluctuation rate.

[0066] B3. If the width fluctuation rate within a layer is less than the set threshold, the average width change gradient of adjacent layers is calculated in the depth direction.

[0067] It's also important to note that the intra-layer width fluctuation rate can help distinguish true crack growth from external interference. For example, if a layer's width varies significantly, it may indicate the presence of foreign matter or measurement error rather than actual crack growth. By calculating the fluctuation rate, these unreliable data points can be identified and addressed, improving overall detection accuracy.

[0068] It can be understood that the intra-layer width fluctuation rate is obtained by comparing the difference between the maximum and minimum values ​​of the intra-layer measured width with the average value of the intra-layer width.

[0069] It should be added that the specific calculation formula for the change gradient is: ,in, Indicates the The depth layer and The average width change gradient of each depth layer is Indicates the depth layer number, , and Respectively represent The depth layer and The average width of the depth layer, Respectively represent The depth layer and The depth coordinate of the depth layer.

[0070] It should also be added that if the width fluctuation rate within a layer is greater than or equal to the set threshold, it is judged as foreign body obstruction, such as gravel stuck in a crack, and an error correction mechanism is required. The error correction mechanism specifically includes: performing sub-layered tomographic scanning on the layer, compressing the layer interval to 1 / 5 of the original setting, using Gaussian filtering to smooth the width data, eliminating outliers and recalculating the average width change gradient of adjacent layers.

[0071] For example, sub-slice scanning identifies the boundaries of tiny foreign bodies by compressing 0.5 mm intervals to 0.1 mm, and Gaussian filtering eliminates high-frequency noise through the filter to retain the true gradient trend.

[0072] Specifically, the thermodynamic parameters include the transverse temperature gradient, the longitudinal temperature decay rate, and the heat diffusion area, and the specific extraction process includes: R1, using the Canny edge detection algorithm to identify the crack edge and extract the continuous contour.

[0073] R2. In the long-wave infrared image, temperature measurement points are selected at equal intervals along both sides of the contour, and the ratio of the average temperature difference on both sides to the actual width of the crack is calculated to obtain the lateral temperature gradient.

[0074] R3. Divide the temperature measurement segments into equal intervals along the extension direction of the contour, and perform linear fitting on the temperature data of each segment to obtain the longitudinal temperature attenuation rate.

[0075] R4. Extract the outer extension area of ​​the crack contour based on the preset temperature boundary, and use the area of ​​the outer extension area of ​​the crack contour as the heat diffusion area.

[0076] In a specific embodiment, the crack contour peripheral expansion area can be extended outward from the crack edge to the boundary where the temperature decays to the ambient temperature ±2°C, and the crack contour is expanded outward by the morphological expansion algorithm until the temperature field meets , the pixel area covered by the extended area is the heat diffusion area. is the current location The temperature value at is the ambient temperature value.

[0077] Understandably, the calculation formula for the longitudinal temperature decay rate is: , represents the longitudinal temperature decay rate, Expressed as temperature change, Indicates the length of the temperature measurement section in the extension direction.

[0078] More specifically, the calculation of the comprehensive risk coefficient includes: E1, calculating the basic risk score based on the crack length and average crack width, and performing nonlinear correction on the score through the crack horizontal development change rate, crack branching density and the change gradient of crack width with depth to obtain a corrected basic risk score.

[0079] It should be noted that calculating the basic hazard score based on crack length and average crack width involves normalizing the crack length and average crack width, and then performing a weighted summation to calculate the basic hazard score. The weights for crack length and average crack width are set based on the crack type. For tensile cracks, the weights for crack length and average crack width can be 0.6 and 0.4, respectively. For shear cracks, the weights for crack length and average crack width can be 0.4 and 0.6, respectively.

[0080] Understandably, when calculating the basic hazard score, the weighting of crack length and average crack width needs to comprehensively consider their impact on the degree of foundation hazard. By collecting a large amount of historical data, such as crack length, average crack width and corresponding actual assessment results of basic hazards under different working conditions and environmental conditions, regression analysis or machine learning algorithms such as linear regression and support vector machines are used to determine the optimal values ​​of crack length weight and average crack width weight to ensure that the weights can reflect the actual impact of the two in different scenarios. After determining the optimal weights, the rationality of the scoring formula is verified with the help of simulation or actual cases to ensure that the basic hazard score accurately reflects the actual degree of hazard. The real-time measured crack length and average crack width are substituted into the formula to calculate the score, providing a scientific basis for foundation maintenance and treatment. This weighting setting comprehensively considers the contribution of crack length and average width to the degree of basic hazard, achieving an accurate assessment of basic hazard.

[0081] It should also be added that the nonlinear correction of the score specifically includes: the maximum-minimum normalization of the fracture horizontal development rate, fracture branching density and fracture width gradient with depth, and the processing results are recorded as 、 and , calculate the correction factor , where the specific calculation formula for the correction coefficient is: , is a natural constant, 、 and They represent the weights corresponding to the set fracture horizontal development rate, fracture branching density, and the fracture width gradient with depth.

[0082] Understandably, in actual projects, the relationship between indicators such as the rate of change in horizontal crack development, crack branching density, and the gradient of crack width with depth, and the final score is often nonlinear. For example, after a certain indicator exceeds a certain threshold, its impact on the score may significantly increase, and linear models cannot capture this change. Using nonlinear functions such as exponential functions can more accurately fit the complex relationship between indicators and scores, making the correction coefficient more realistic. This formula can comprehensively consider the nonlinear corrections to the basic hazard score caused by multiple factors, improving the accuracy and reliability of the assessment results and providing a more precise basis for engineering decision-making.

[0083] It is also necessary to add that 、 and They are used to measure the relative importance of the horizontal crack development rate, crack branching density, and crack width gradient with depth to the modified basic hazard score. The larger the weight value, the more significant the impact of the corresponding factor on the correction coefficient, for example When it is larger, it indicates that the crack level development change rate plays a more critical role in the risk score correction. The setting of these weights needs to be determined in combination with historical engineering data and statistical analysis. The weights can be trained and optimized based on the actual risk assessment results through regression analysis, least squares optimization, etc. It can accurately reflect the actual degree of danger under the combined effect of various factors. After determining the weight, it is necessary to verify it through multiple sets of measured data to ensure that under different working conditions, the correction coefficient can reasonably reflect the influence of various factors. For example, when the crack branch density is high and When it is larger, Should be significantly increased to reflect the increased risk.

[0084] E2. Calculate the thermodynamic score based on the thermodynamic parameters and combine it with the modified basic risk score according to the preset weight to obtain the current comprehensive risk factor.

[0085] Specifically, the thermodynamic parameters are processed in the same way as the correction coefficient processing principle to obtain the thermodynamic score.

[0086] Understandably, the weights of the thermodynamic score and the basic hazard score are determined by combining the analytic hierarchy process with data-driven methods such as the entropy weight method. Specifically, a judgment matrix can be constructed through expert experience to quantitatively evaluate the relative importance of geometric morphology and thermodynamic anomalies to structural safety, such as a basic score weight of 0.6 and a thermodynamic score of 0.4. Historical accident data can then be used for regression verification and automatic optimization of the weight ratio.

[0087] S4. Generate a graded thermal map based on the spatial distribution of the current comprehensive risk factor, and trigger a graded warning signal in combination with the expansion trend to generate a crack detection report.

[0088] It should be added that the surface of the asphalt layer is gridded to obtain various collection areas, and steps S1 to S3 are repeated for each collection area to obtain the comprehensive hazard factor of each collection area. The comprehensive hazard factor of each collection area is then mapped to the electronic map of the construction area to generate a graded thermal map.

[0089] The embodiment of the present invention calculates the current comprehensive risk factor by constructing a weighted scoring model based on the geometric morphology of cracks and thermodynamic dynamic parameters, thereby realizing a quantitative risk assessment of cracks and making up for the defect of relying solely on qualitative assessment of temperature differences. At the same time, by generating a spatial distribution thermal map and a crack detection report, the risk level distribution is intuitively quantified, significantly reducing the engineering safety hazards caused by assessment deviations.

[0090] Furthermore, triggering the graded warning signal includes: matching the comprehensive risk factor with the associated comprehensive risk factors under each pre-set warning level to obtain a matching warning level, which is used as the initial warning level.

[0091] Obtain the comprehensive risk factor sequence of the current and historical detection periods, and calculate the risk factor change rate of adjacent periods.

[0092] The number of detection cycles in which the risk factor continuously increases in the sequence is counted, and the ratio of the continuous increase to the total number of detection cycles is calculated.

[0093] When the risk factor change rate exceeds the first preset warning threshold, or the continuous growth rate exceeds the second preset warning threshold, the initial warning level is raised to the next warning level, triggering the warning signal of the next warning level.

[0094] It should also be added that the crack detection report automatically integrates multispectral images, graded thermal maps, multi-band image data and graded warning signals.

[0095] The embodiment of the present invention calculates the current comprehensive risk factor by constructing a weighted scoring model based on the geometric morphology of cracks and thermodynamic dynamic parameters, thereby realizing a quantitative risk assessment of cracks and making up for the defect of relying solely on qualitative assessment of temperature differences. At the same time, by generating a spatial distribution thermal map and a crack detection report, the risk level distribution is intuitively quantified, significantly reducing the engineering safety hazards caused by assessment deviations.

[0096] See also Figure 2 As shown, the present invention provides a crack detection system for highway construction, which includes: a multi-spectral thermal imaging acquisition module, a crack interference identification and elimination module, a crack risk analysis module and a crack report generation terminal.

[0097] Among them, the crack interference identification and elimination module is connected to the multi-spectral thermal imaging acquisition module and the crack risk analysis module respectively, and the crack risk analysis module is connected to the crack report generation terminal.

[0098] The multispectral thermal imaging acquisition module uses a multispectral thermal imaging sensor to synchronously collect multi-band image data of the asphalt layer surface and perform registration and fusion.

[0099] The crack interference identification and elimination module eliminates false crack interference data based on the correlation analysis of edge continuity features and thermal radiation fluctuation features, and obtains effective multi-band image data.

[0100] The crack risk analysis module extracts the geometric parameters and thermodynamic parameters of the cracks from the effective multi-band image data and calculates the current comprehensive risk factor through a weighted scoring model.

[0101] The crack report generation terminal generates a graded thermal map based on the current spatial distribution of the comprehensive hazard factor, and triggers a graded warning signal based on the expansion trend to generate a crack detection report.

[0102] The preset parameters in the above formula are set by those skilled in the art according to actual conditions.

[0103] The above embodiments may be implemented in whole or in part through software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0104] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0105] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0106] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0107] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A crack detection method for highway construction, characterized by: The method comprises the following steps: S1. Use a multispectral thermal imaging sensor to synchronously collect multi-band image data of the asphalt layer surface and perform registration and fusion. The multi-band includes visible light band, near infrared band and long-wave infrared band; S2. Based on the correlation analysis of edge continuity characteristics and thermal radiation fluctuation characteristics, false crack interference data is eliminated and effective multi-band image data is obtained; S3. Extracting the geometric parameters and thermodynamic parameters of the cracks from the effective multi-band image data and calculating the current comprehensive risk factor using a weighted scoring model; S4. Generate a graded thermal map based on the spatial distribution of the current comprehensive risk factor, trigger a graded warning signal based on the expansion trend, and generate a crack detection report; The specific elimination process of eliminating false crack interference data includes: The visible light band image is grayscaled, the sliding window size is set based on the minimum apparent width of the crack, the texture roughness of each pixel neighborhood is traversed and calculated, and the areas with texture roughness below the preset threshold range are marked as low roughness areas; In a sequence of thermal radiation images synchronously collected in the near-infrared band, the time-series fluctuation data of the thermal radiation intensity of the low-roughness area is extracted, and its standard deviation and mean offset are calculated. At the same time, in the long-wave infrared band image, the deviation value of the temperature distribution of the shadow area and the corresponding brightness distribution of the visible light image is calculated; The edge continuity features of the detection area are extracted from the visible light band image, combined with the thermal radiation intensity fluctuation characteristics of the near-infrared band to construct the texture-thermal radiation correlation matrix, and the correlation coefficient between the two is calculated; Interference determination is performed on the standard deviation, mean offset, deviation value and correlation coefficient using a preset interference determination rule, and image data in the visible light, near infrared and long-wave infrared bands are simultaneously deleted for the area determined to be interfered with; The specific determination process of the interference determination includes: if the standard deviation of the thermal radiation intensity is lower than the lower limit of the standard deviation reference range of the crack characteristic model and its mean offset exceeds the preset offset threshold, it is determined that there is surface attachment interference; If the deviation value exceeds a preset adjustment threshold, it is determined that optical shadow artifact interference exists; If the correlation coefficient is lower than a preset correlation coefficient threshold, it is determined that there is stain or foreign matter interference.

2. A crack detection method for highway construction according to claim 1, characterized in that: The specific execution process of step S1 includes: The asphalt layer surface image is collected synchronously by using a multispectral sensor in the visible light band, near infrared band and long-wave infrared band; The image registration algorithm is used to spatially align and fuse multi-band images to generate a fused image containing temperature gradient distribution and texture features; The geographic coordinates of each acquisition area are marked, and the fused image is bound to the coordinate information.

3. The crack detection method for highway construction according to claim 1, characterized in that: The geometric parameters of the cracks include crack length, average crack width, crack horizontal development rate, crack branching density, and crack width gradient with depth. The specific extraction process includes: The Canny edge detection algorithm is used to identify the crack center in the visible light band image. The longest continuous pixel distance along the crack centerline is measured and converted into the actual length based on the image resolution to be used as the crack length. Take multiple sampling points perpendicular to the crack centerline and calculate the average value as the average crack width; The width is measured segmentally along the crack development direction, and the slope of the width versus extension distance is fitted by linear regression to be used as the horizontal development rate of the crack width. The number of crack bifurcation points is counted from the visible light band image and compared with the area of ​​the crack region to obtain the crack branch density; The corresponding crack cross-sectional profile in the visible light band image is extracted, and then the width is measured layer by layer along the depth direction, and its gradient with depth is calculated.

4. A crack detection method for highway construction according to claim 3, characterized in that: The specific calculation process of the change gradient includes: Cross sections are taken at equal intervals along the extension direction of the crack centerline, and layers are formed at dynamic intervals along the depth direction on each cross section; Measure the maximum and minimum width values ​​in each layer and calculate the width fluctuation rate within the layer; If the width fluctuation rate within a layer is less than the set threshold, the average width change gradient of adjacent layers is calculated in the depth direction.

5. The crack detection method for highway construction according to claim 1, characterized in that: The thermodynamic parameters include the transverse temperature gradient, the longitudinal temperature decay rate, and the heat diffusion area, and the specific extraction process includes: Use the Canny edge detection algorithm to identify crack edges and extract continuous contours; In the long-wave infrared image, temperature measurement points are selected at equal intervals along both sides of the contour, and the ratio of the average temperature difference on both sides to the actual width of the crack is calculated to obtain the transverse temperature gradient. The temperature measurement segments are divided into equal intervals along the extension direction of the contour, and the longitudinal temperature attenuation rate is obtained by linear fitting of the temperature data of each segment; The peripheral extension area of ​​the crack contour is extracted based on the preset temperature boundary, and the area of ​​the peripheral extension area of ​​the crack contour is used as the heat diffusion area.

6. A crack detection method for highway construction according to claim 3, characterized in that: The calculation of the current comprehensive risk factor includes: The basic hazard score is calculated based on the crack length and average crack width, and the score is nonlinearly corrected by the crack horizontal development change rate, crack branching density, and the crack width gradient with depth to obtain the corrected basic hazard score. The thermodynamic score is calculated based on the thermodynamic parameters and is combined with the modified basic risk score according to the preset weights to obtain the current comprehensive risk factor.

7. The crack detection method for highway construction according to claim 1, characterized in that: The triggering of the graded warning signal includes: The comprehensive risk factor is matched with the associated comprehensive risk factors under each pre-set warning level to obtain the matching warning level, which is used as the initial warning level; Obtain the comprehensive risk factor sequence of the current and historical detection periods, and calculate the risk factor change rate of adjacent periods; Counting the number of detection cycles in which the risk factor continuously increases in the sequence, and calculating the ratio of the continuous increase to the total number of detection cycles; When the risk factor change rate exceeds the first preset warning threshold, or the continuous growth rate exceeds the second preset warning threshold, the initial warning level is raised to the next warning level, triggering the warning signal of the next warning level.

8. A crack detection system for highway construction, configured to execute the steps of a crack detection method for highway construction according to any one of claims 1 to 7, characterized in that: The system includes the following modules: The multispectral thermal imaging acquisition module uses a multispectral thermal imaging sensor to synchronously collect multi-band image data of the asphalt layer surface and perform registration and fusion; The crack interference identification and elimination module eliminates false crack interference data based on the correlation analysis of edge continuity features and thermal radiation fluctuation features, and obtains effective multi-band image data; The crack risk analysis module extracts the geometric and thermodynamic parameters of the cracks from the effective multi-band image data and calculates the current comprehensive risk factor through a weighted scoring model; The crack report generation terminal generates a graded heat map based on the current spatial distribution of the comprehensive hazard factor, and triggers a graded warning signal in combination with the expansion trend to generate a crack detection report.

Images