Cast-in-place floor crack construction control method and system based on image processing

Abstract

The invention relates to the technical field of floor pouring construction monitoring, in particular to a cast-in-place floor crack construction control method and system based on image processing. According to the method, the three types of key defects including the honeycomb defect, the layering defect and the crack defect are identified in a combined mode, the area proportion, the regional dispersion degree and the direction weighting influence degree of the three types of key defects are quantified, multi-dimensional and refined evaluation of the cast-in-place floor slab structure state is achieved, and the accuracy and comprehensiveness of follow-up pouring section crack risk prediction are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the floor pouring construction monitoring technical field, and in particular to a cast-in-place floor crack construction control method and system based on image processing. BACKGROUND

[0002] The cast-in-place concrete floor is a key horizontal load-bearing component of modern building structure, and its pouring construction quality is directly related to the structural safety and durability of the overall building. During the construction process, due to the influence of factors such as material mix ratio, pouring speed, and vibration intensity, honeycomb, delamination, and cracks and other defects are easily generated on the floor surface.

[0003] The existence of defects will affect the durability and safety of the floor section. Therefore, in order to ensure the engineering quality, the floor state needs to be monitored in real time during the pouring construction stage, and the construction parameters need to be adjusted in time and effectively according to the monitoring results.

[0004] In view of the influence of pouring speed and vibration intensity, the following defects exist in the current monitoring of the construction quality of the cast-in-place floor: it is difficult to convert the geometric characteristics of static and apparent defects into accurate prediction of future dynamic crack evolution risk, resulting in a lack of basis for forward-looking decision-making in construction quality control. It is impossible to dynamically regulate the pouring speed and vibration intensity based on real-time quality feedback, and there is a lack of closed-loop control for active prevention, which leads to mismatch of construction process parameters and easily causes quality defects such as over-vibration segregation or under-vibration incompactness. SUMMARY

[0005] The purpose of the present application is to overcome the defects of the prior art, and to provide a cast-in-place floor crack construction control method and system based on image processing, in order to solve the problem that there is no correlation between static defect quantitative evaluation and dynamic crack risk forward-looking prediction in the prior art, and construction parameters cannot be adjusted based on real-time quality feedback, resulting in process parameter mismatch and quality defects.

[0006] The technical solution adopted by the present application to solve its technical problems is: a cast-in-place floor crack construction control method based on image processing, comprising: collecting surface images of key areas of the current set floor section, and identifying honeycomb defects, delamination defects, and crack defects in the surface images.

[0007] The area ratio of the honeycomb defects and the area dispersion of the delamination defects are calculated; the width and direction of the crack defects are obtained, and the total influence degree is determined accordingly.

[0008] The area ratio, area dispersion, and total influence degree are input into a preset defect degree analysis model, and the crack risk value of subsequent pouring is output.

[0009] Construction parameter adjustment instructions including pouring speed and vibration intensity are generated according to the crack risk value.

[0010] The construction parameter adjustment command is sent to the pouring equipment to pour the next floor slab segment.

[0011] After the next floor slab has solidified, its surface image is acquired and a new crack risk value is determined. If the new crack risk value is less than or equal to the preset safety threshold, the current construction parameters are maintained; otherwise, a new construction parameter adjustment instruction is generated.

[0012] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention achieves a multi-dimensional and refined assessment of the state of cast-in-place floor slab structure by jointly identifying three types of key defects: honeycomb defects, delamination defects and crack defects, and quantifying their area proportion, regional dispersion and directional weighted influence, which significantly improves the accuracy and comprehensiveness of crack risk prediction in subsequent pouring sections.

[0013] (2) The present invention calculates the risk correction coefficient based on the predicted crack risk value, and realizes the nonlinear coordinated control of pouring speed and vibration intensity through piecewise function relationship, dynamically balancing quality and efficiency, and making the static defect quantitative assessment and dynamic crack risk prospective prediction closely related.

[0014] (3) In high-risk conditions, the present invention reduces the pouring speed and increases the vibration intensity to enhance the density of concrete and inhibit the development of early plastic cracks; in low-risk conditions, the pouring speed is increased and the vibration intensity is reduced to prevent excessive vibration from causing concrete segregation, while saving energy. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of the control method of the present invention.

[0017] Figure 2 This is a flowchart for determining whether correction is needed according to the present invention.

[0018] Figure 3 The flowchart for identifying crack defects in this invention Figure 4 This is a schematic diagram of the system module connections of the present invention. Detailed Implementation

[0019] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention. Furthermore, it should be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale.

[0020] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use. Techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and apparatus should be considered part of the specification.

[0021] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0022] Please see Figure 1 As shown, a construction control method for cracks in cast-in-place floor slabs based on image processing includes: acquiring surface images of key areas of currently solidified floor slab sections, and identifying honeycomb defects, delamination defects, and crack defects in the surface images.

[0023] During concrete pouring, excessively fast pouring speed or insufficient vibration intensity can prevent air from being fully expelled from the concrete, making it difficult for mortar to fill the gaps between aggregates and forming localized clusters of voids, i.e., honeycomb defects.

[0024] The presence of honeycomb defects can significantly reduce the effective load-bearing area of ​​components and cause stress concentration when under stress, which is a common origin of structural cracks.

[0025] When the pouring speed is uneven or the vibration intensity is insufficient along the depth direction, the surface concrete will be dense due to sufficient vibration, while the deep concrete will be loose due to insufficient vibration intensity. A density interface is formed between the loose area and the dense surface layer. Over time, the interface will peel off due to the difference in shrinkage rate, forming a delamination defect.

[0026] Delamination defects create potentially weak interfaces within the concrete, severely weakening the interlayer bond and shear strength. Under subsequent loads or temperature deformation, these defects can easily develop into peeling, flaking, or deep cracks, posing a highly insidious safety hazard.

[0027] Pouring too fast or too slow, or abnormal vibration intensity, can all cause shrinkage stress in concrete during the setting process. When the stress exceeds the tensile strength of the concrete, cracks will form.

[0028] The presence of cracks directly weakens the tensile strength and overall load-bearing capacity of cast-in-place floor slabs, and accelerates the degradation of structural durability.

[0029] Cellular defects, delamination defects, and crack defects provide complementary and non-substitutable state information from the three dimensions of material uniformity, continuity, and integrity, respectively, ensuring targeted control of floor slab crack risks.

[0030] In the implementation of the above scheme, the process of collecting surface images of key areas of the currently solidified floor slab segment is as follows: the mid-span stress zone, the area around the beam-column joint, and the area on both sides of the construction joint of the currently solidified floor slab segment are taken as key areas.

[0031] The mid-span stress zone, the area around beam-column joints, and the areas on both sides of construction joints are high-incidence areas for defects in cast-in-place floor slabs. These three areas have a far greater impact on the structural safety of the floor slab and the risk of subsequent cracking than non-critical areas.

[0032] Among them, the mid-span stress zone is the core area with the largest bending moment when the cast-in-place floor slab bears vertical loads, and it is prone to cracks due to load concentration; the area around the beam-column joint is the connection between the floor slab and the beam-column, where there is a combined effect of horizontal and vertical forces, which is prone to stress concentration, resulting in honeycomb defects and delamination defects; the area on both sides of the construction joint is the interface between the old and new concrete pouring, which is a naturally formed weak part of the construction process, and is prone to delamination defects and crack defects.

[0033] By acquiring images of the mid-span stress zone, the area surrounding the beam-column joint, and the areas on both sides of the construction joint, we can systematically cover the bending stress sensitive area, the shear and structural sensitive area, and the construction process sensitive area. While ensuring risk coverage, we can avoid redundant data interference in non-critical areas and improve the efficiency and accuracy of defect identification and risk assessment.

[0034] Adjust the image acquisition device to an angle where the optical axis is perpendicular to the surface of the currently solidified floor slab section; the image acquisition device scans the area in the mid-span stress zone, the area around the beam-column joint, and the area on both sides of the construction joint according to the preset path scanning scheme.

[0035] It should be noted that, according to the principles of perspective geometry, aligning the optical axis of the image acquisition device perpendicular to the surface of the currently solidified floor slab is to eliminate perspective distortion and avoid data distortion.

[0036] In the implementation of the above scheme, the process of establishing the path scanning scheme is as follows: obtain the rectangular boundary dimensions of the mid-span stress zone, the annular boundary dimensions of the area surrounding the beam-column joint, and the strip boundary dimensions of the areas on both sides of the construction joint.

[0037] For the mid-span stress zone, a grid-like scanning path covering the entire rectangular boundary is generated, and the grid-like scanning path is composed of interwoven horizontal and vertical paths that are perpendicular to each other.

[0038] The mid-span stress zone is typically a regular rectangle with a large area of ​​uniform stress. Defects can appear anywhere within this rectangular uniform stress zone without a clear direction.

[0039] Within a regular rectangular uniform stress area, the total length and acquisition time of the grid-like scanning path are generally superior to other complex scanning paths. At the same time, it can also achieve no dead-angle coverage of the rectangular uniform stress area, ensuring that no defects are missed.

[0040] For the area surrounding the beam-column joint, a radial scanning path is generated with the center of the beam-column as the reference, and the radial scanning path is evenly distributed in the radial and circumferential directions.

[0041] The area surrounding the beam-column joint is a centrally symmetrical annular region centered on the beam and column. A radial scanning path can most naturally cover this area, and the path planning logic is simple and efficient.

[0042] Meanwhile, since stress and cracks typically develop radially from the center outwards, a radial scanning path can be perpendicular to or at a large angle to the direction of the most prevalent crack development in the area surrounding the beam-column joint. This allows for the clearest capture of the complete shape and width of the crack during imaging, avoiding missed detections or measurement distortions caused by the path being parallel to the crack, and also preventing over-sampling in the central area or under-sampling in the peripheral area.

[0043] For the areas on both sides of the construction joint, a bow-shaped scanning path is generated parallel to the extension direction of the construction joint, and the bow-shaped path is symmetrically distributed on both sides of the construction joint.

[0044] Construction joints are typical linear bands, with risk areas distributed within a certain width on both sides of the joint.

[0045] The scanning direction of the bow-shaped path is basically perpendicular to the extension direction of the potential band-like defect. This perpendicularly intersecting sampling method can most effectively highlight the image features of the defect, ensuring that it is captured clearly and completely, and greatly improving the recognition rate.

[0046] The symmetrical distribution of the bow-shaped path on both sides of the construction joint ensures that the quality of both the old and new concrete is inspected equally.

[0047] By using different optimized scanning paths for different key areas, we ensure that the acquired images are high-quality data that best reflects the typical defect characteristics of the key area, thereby guaranteeing the accuracy of subsequent defect identification.

[0048] The generated scanning path is converted into motion control commands for the image acquisition device, forming a complete path scanning scheme.

[0049] Please see Figure 2 and Figure 3 As shown, in the implementation of the above scheme, the process of identifying honeycomb defects, delamination defects and crack defects in the surface image is as follows: the acquired surface image is divided into multiple candidate regions, and the following screening is performed on each candidate region: the ratio of the area of ​​the candidate region to the area of ​​the minimum bounding rectangle is calculated, and the change in the area of ​​the candidate region before and after the morphological closing operation is calculated; if the ratio is less than the first reference benchmark and the change is greater than zero, then the candidate region is identified as a honeycomb defect.

[0050] A regular region has an area whose ratio to its smallest bounding rectangle is close to 1. However, a honeycomb defect is composed of a large number of irregular, discrete voids and protrusions, whose shapes are extremely irregular and cannot fill its bounding rectangle.

[0051] For a dense region, the area remains almost unchanged before and after the closing operation. However, for a honeycomb defect, the numerous small holes and gaps inside are filled by the closing operation, resulting in a significant increase in the area of ​​the region after the operation.

[0052] A cellar defect is identified only when the ratio of the candidate region area to the area of ​​the minimum circumscribed rectangle is less than the first reference benchmark and the change is greater than zero.

[0053] Calculate the aspect ratio of the minimum bounding rectangle of the candidate region. If the aspect ratio is greater than the second reference benchmark, it is determined to be a strip region. Extract the grayscale profile curve along its short side and calculate the ratio of the curve trough depth to the background grayscale standard deviation. If the ratio is greater than the third reference benchmark, the candidate region is identified as a layered defect.

[0054] Layered defects typically appear in images as thin, shallow stripes, with the longer side of the smallest bounding rectangle being much larger than the shorter side. By comparing the aspect ratio and a second reference, the target can be quickly identified from a large pool of candidate regions, focusing on areas with elongated shapes.

[0055] The ratio of the curve trough depth to the standard deviation of the background grayscale reflects the grayscale difference between the delamination defect band and the normal concrete area, and its significance relative to the natural grayscale fluctuations of the normal concrete area itself. This allows us to determine whether the low-grayscale bands in the image are genuine delamination defects or accidental grayscale fluctuations in normal concrete caused by slight changes in aggregate distribution or lighting.

[0056] It should be noted that the first, second, and third reference benchmarks can be determined comprehensively based on the type of concrete, structural location, and imaging environment.

[0057] In this invention, for example, for concrete with aggregate particle size of 20-30mm, located at beam-column joints and imaged indoors, the first reference reference can be set to 0.5, the second reference reference to 3, and the third reference reference to 2.

[0058] For concrete with aggregate particle size of 10-20mm, located at construction joints and for outdoor imaging, the first reference benchmark can be set to 0.6, the second reference benchmark to 5, and the third reference benchmark to 3.

[0059] Please see Figure 3 As shown, the candidate region is skeletonized to obtain the central skeleton line, which is used to identify crack defects. The specific process is as follows: determine whether there are branch points in the central skeleton line; if there are branch points, perform the following operation: extract all branch points connecting three or more skeleton lines from the central skeleton line.

[0060] During their development, cracks may encounter obstacles such as aggregates and reinforcing bars, or may form two main categories due to changes in their own stress state: simple linear cracks and complex dendritic cracks.

[0061] For different linear and dendritic cracks, the most suitable quantitative assessment method is adopted to ensure that the cracks can be accurately quantified and identified, thereby improving the identification accuracy.

[0062] Using each branch point as a boundary, the central skeleton line is divided into multiple skeleton line segments; the ratio of the total length of each skeleton line segment to the distance between its first and last endpoints is taken as the meandering degree; the maximum value among all meandering degrees is taken as the final meandering degree index of the candidate region.

[0063] For cracks with branches, the ratio of the total length of the central skeleton line to the distance between its first and last endpoints will be severely distorted due to the presence of branches. Therefore, it is necessary to decompose it into multiple skeleton line segments and then find the most tortuous segment as the evaluation criterion.

[0064] If the final meandering index is greater than the fourth reference benchmark, the candidate region is identified as a crack defect.

[0065] The fundamental difference between cracks and non-cracked linear regions lies in their tortuosity. Cracks have highly irregular and meandering paths, resulting in a tortuosity value far greater than 1. This tortuosity, a core indicator, can accurately distinguish between genuine, naturally formed cracks and regular, straight artificial or processed traces.

[0066] Otherwise, calculate the ratio of the total length of the central skeleton line to the distance between its first and last endpoints. If the ratio is greater than the fourth reference datum, then the candidate region is identified as a crack defect.

[0067] The fourth reference benchmark can be determined according to the resolution of the image acquisition device. For example, for an image acquisition device with a resolution of 1920×1080 pixels, the fourth reference benchmark can be set to 2.5 to avoid missing short branch cracks; for an image acquisition device with a resolution of 1280×720 pixels, the fourth reference benchmark can be set to 3 to avoid noise misjudgment.

[0068] The calculation of the area proportion of cellular defects and the regional dispersion of layered defects is specifically as follows: the total area of ​​all cellular defects is counted, and the ratio of this area to the total area of ​​the key area is calculated to obtain the area proportion of cellular defects.

[0069] Obtain the geometric center coordinates of each layer of defects, and transform all geometric center coordinates to the same floor slab actual coordinate system.

[0070] If there is no stratification defect, the regional dispersion is set to zero; if there is a unique stratification defect, the Euclidean distance between its geometric center coordinates and the geometric center of the floor slab is used as the regional dispersion.

[0071] When there are no stratification defects, the regional dispersion is naturally zero. Treating this case separately avoids meaningless subsequent mathematical calculations.

[0072] The risk of a single defect stems primarily from the isolation and criticality of its location, rather than from regional dispersion. The greater the Euclidean distance between the geometric center coordinates of a single layered defect and the geometric center of the floor slab, the further the layered defect is from the structural center, potentially located in sensitive areas such as edges, and its potential for hazard is extremely high.

[0073] By assigning a non-zero risk value to a single defect, information loss is avoided, making the feature of regional dispersion valuable for evaluation in all cases, thereby ensuring the accuracy and reliability of the final construction control.

[0074] If there are non-unique layered defects, calculate the average of the X and Y coordinates of all geometric centers to obtain the set center point of the layered defect distribution.

[0075] Calculate the Euclidean distance from each geometric center coordinate to the set center point, and use the average of all Euclidean distances as the regional dispersion. The larger the regional dispersion, the more widespread and dispersed the layered defects are in the floor slab, which usually means there are systemic problems in the construction process; the smaller the dispersion, the more concentrated the layered defects are, and the relatively controllable risk range.

[0076] The process of obtaining the width and direction of the crack defect and determining the total influence is as follows: select multiple measurement points at equal intervals along each central skeleton line, and draw a perpendicular line segment of the central skeleton line at each measurement point.

[0077] For each crack defect, the average physical width of each measurement point is taken as the representative width of that crack defect.

[0078] The physical width is obtained by converting the pixel length of the vertical line segment in the image into actual physical units through image analysis and in combination with the calibration parameters of the image acquisition device.

[0079] The direction of the line connecting the first and last endpoints of the central skeleton line is taken as the overall direction of the crack defect; the absolute value of the angle between the overall direction and the reference direction is calculated as the direction influence factor.

[0080] Cracks perpendicular to or at a large angle to the principal tensile stress direction cause the most severe weakening of the structural bearing capacity. Therefore, in this invention, by way of example, the aforementioned reference direction can be set as a direction perpendicular to the principal force direction of the floor slab.

[0081] Multiply the width of the crack defect by the direction influence factor to obtain the individual influence degree of the crack defect; sum all the individual influence degrees as the total influence degree of the crack defect.

[0082] The impact of cracks on the structural safety of floor slabs depends on their width and the relative relationship between their extension direction and the direction of the principal tensile stress in the structure.

[0083] When the crack direction is perpendicular to the principal tensile stress direction, it will completely cut off the stress path and severely weaken the load-bearing capacity of the component; when the crack direction is parallel to the principal tensile stress direction, the direct impact on the load-bearing capacity is small, and it may mainly affect durability or aesthetics.

[0084] By introducing a directional influence factor, a comprehensive assessment from both width and direction dimensions is achieved, ensuring that the system can identify that a crack that is not wide but has a dangerous direction may require more aggressive intervention and control than a crack that is wider but has a safe direction.

[0085] Input the area ratio, regional dispersion, and total impact into the preset defect degree analysis model, and output the crack risk value for subsequent pouring.

[0086] In the implementation of the above scheme, the process of constructing the defect degree analysis model is as follows: collect the area ratio, regional dispersion, total impact and crack risk value of the solidified floor slab sections in historical construction projects to form a dataset.

[0087] The dataset is divided into training, validation, and test sets according to a set ratio. The area proportion, regional dispersion, and total impact in the training set are used as input features, and the crack risk value is used as the target variable. The gradient boosting tree model is used for iterative training. By minimizing the loss function, a nonlinear mapping relationship between the input features and the target variable is established, and an initial defect degree analysis model is constructed.

[0088] The initial defect severity analysis model is used to predict the crack risk value of the test set. The mean square error is then used to adjust and optimize the initial defect severity analysis model, and the final defect severity analysis model is output.

[0089] The mean squared error is used to measure the average squared difference between the crack risk value predicted by the model and the actual crack risk value; the smaller the value, the higher the prediction accuracy and the better the stability of the model.

[0090] The initial defect severity analysis model is adjusted and optimized as follows: When the mean squared error of the model's prediction on the validation set is greater than the set mean squared error threshold, the hyperparameters of the gradient boosting tree model are adjusted, including but not limited to the learning rate, the maximum depth of the decision tree, and the minimum number of samples in the leaf nodes; then the model is retrained using the adjusted hyperparameters, and its mean squared error is evaluated again on the validation set; this process is repeated until the mean squared error of the model on the validation set is less than the set mean squared error threshold. At this point, the model is finally evaluated on the test set. If its performance meets the requirements, the output is the final defect severity analysis model.

[0091] This invention acquires surface images of already solidified floor slab sections and precisely quantifies three key features: the area ratio of honeycomb defects, the regional dispersion of delamination defects, and the total impact of crack defects. This enables accurate prediction of future crack risk values ​​from multi-dimensional defect states, significantly improving the prediction accuracy of crack risk in subsequent poured floor slab sections. This provides a core decision-making basis for generating scientific and precise construction parameter adjustment instructions, effectively ensuring the construction quality and structural safety of cast-in-place floor slabs.

[0092] The process of generating construction parameter adjustment instructions including pouring speed and vibration intensity based on crack risk value is as follows: obtain the initial pouring speed and initial vibration intensity of the currently solidified floor slab segment during construction, and use them as the benchmark pouring speed and benchmark vibration intensity.

[0093] Vibration intensity is essentially a physical quantity characterizing the mechanical vibration energy input to a unit volume of concrete. A single parameter cannot fully describe this energy input process; it needs to be calculated comprehensively using the power of the vibrating equipment, vibration frequency, amplitude, and duration of action. The specific calculation formula is as follows: ;in, This is the initial vibration strength. For the power of the vibrator, The vibration frequency, Let be the amplitude. The square of the amplitude reflects the relationship between the vibrational energy and the square of the amplitude. The average duration of vibration. This is the equipment efficiency coefficient. The volume to be covered by vibration.

[0094] vibrator power Vibration frequency ,amplitude The average vibration duration can be obtained directly from the vibrating equipment. The equipment efficiency coefficient can be obtained from the construction records during the construction of the already solidified floor slab section. This value can be provided by the equipment manufacturer, and its range is usually 0.6-0.9; in this invention, 0.7 is used. (Using the formula...) The calculated compaction and covering volume is as follows: The effective radius of action is given and can be obtained from the equipment's technical manual. The insertion depth is specified and can be obtained from the construction records. Pi is the mathematical constant of a circle.

[0095] After obtaining the vibration intensity of each vibration point using the above calculation formula, the average value of the vibration intensity of all vibration points of the entire currently solidified floor slab section is taken as the initial vibration intensity of the floor slab section.

[0096] The initial pouring speed can be directly read from the historical data of the pouring speed set during the construction of the current solidified floor slab section from the concrete pouring equipment. The average pouring speed of all vibration points is used as the initial pouring speed of the current solidified floor slab section.

[0097] The ratio of the crack risk value of the current solidified floor slab segment to the benchmark risk value is used as the risk correction coefficient.

[0098] The baseline risk value can be determined based on the type of floor slab to which the currently solidified floor slab segment belongs. For example, for ordinary reinforced concrete cast-in-place slabs with a thickness of 100-120mm commonly used in residential buildings, the baseline risk value can be set at 0.35; for reinforced concrete cast-in-place slabs with a thickness of 150-180mm used in commercial buildings to support heavy equipment or people, the baseline risk value can be set at 0.5. Based on the risk correction coefficient, the adjusted pouring speed is determined according to the first functional relationship, and the adjusted vibration intensity is determined according to the second functional relationship. The first and second functional relationships are piecewise functional relationships, and the adjusted pouring speed decreases monotonically with the increase of the risk correction coefficient, while the adjusted vibration intensity increases monotonically with the increase of the risk correction coefficient.

[0099] The aforementioned piecewise functional relationship is constructed based on the risk correction coefficient. With the predetermined segmentation threshold The comparison results show that this invention achieves the core principle of adaptive control by reducing the speed of strong vibrations to ensure quality during high-risk situations and increasing the speed of weak vibrations to improve efficiency during low-risk situations through piecewise functional relationships.

[0100] The segmented adjustment is specifically reflected in: when When this condition is indicated, it signifies a high-risk working condition, and the control objective at this time is to improve the pouring quality. This can be specifically achieved through a function. The adjusted pouring speed was calculated. By reducing the pouring speed, the amount of concrete poured in per unit time is reduced, allowing sufficient time for vibration. This is achieved through a function. The adjusted vibration strength was calculated. By increasing the vibration energy, the density of concrete is improved, thus bridging potential defects.

[0101] when When this condition is met, it indicates a low-risk working condition, and the control objective at this time is to optimize construction efficiency. This can be specifically achieved through a function. The adjusted pouring speed was calculated. While ensuring quality, appropriately increase speed to improve efficiency. This can be achieved through functions. The adjusted vibration strength was calculated. It saves energy while preventing concrete segregation caused by excessive vibration.

[0102] when At the same time, the benchmark pouring speed and benchmark vibration intensity remain unchanged.

[0103] in, As a benchmark pouring speed, The benchmark vibration intensity is used. This represents the critical point at which the risk of cracking changes from acceptable to requiring intervention, and is typically set to 1; however, it can also be adjusted according to the specific quality requirements of the project, for example, for projects with strict quality control. A value of 0.8 is acceptable; for projects where construction efficiency is more important, 1.2 is acceptable.

[0104] The adjusted pouring speed is limited between the preset lower and upper speed thresholds to obtain the final pouring speed; the vibration intensity is handled in the same way.

[0105] Pouring too slowly can cause the concrete to set prematurely during the pouring process, creating artificially created weak points; pouring too quickly can lead to a series of problems such as concrete segregation and excessive lateral pressure on the formwork.

[0106] Insufficient vibration intensity will fail to ensure the compactness of concrete, making it difficult to effectively remove air bubbles and thus forming defects such as honeycomb; excessive vibration will lead to severe segregation of concrete, sinking of coarse aggregate, and may damage the formwork or the positioning of reinforcing bars.

[0107] Therefore, by limiting the adjusted pouring speed and vibration intensity, the feasibility of the construction process is ensured, and the safe operation of the equipment is guaranteed.

[0108] Generate construction parameter adjustment instructions containing the final pouring speed and final vibration intensity; send the construction parameter adjustment instructions to the pouring equipment to pour the next floor slab segment.

[0109] The pouring speed and vibration intensity together determine the compaction quality of concrete. For example, when the pouring speed is increased, the amount of concrete poured into the formwork per unit time increases, requiring a simultaneous increase in vibration intensity and a longer vibration time to ensure that the newly added concrete is fully compacted. Conversely, when the pouring speed is reduced, the amount of concrete poured into the formwork per unit time decreases. If the original vibration intensity is maintained, it may lead to excessive vibration of local concrete, resulting in segregation.

[0110] By simultaneously, collaboratively, and in reverse adjusting the pouring speed and vibration intensity, the pouring speed and vibration intensity are dynamically matched. This effectively avoids over-vibration and under-vibration while ensuring that the concrete is fully compacted, thereby achieving the optimal balance between quality and efficiency. This is the key to the intelligent and precise control achieved by this invention.

[0111] After the next floor slab has solidified, its surface image is acquired and a new crack risk value is determined. If the new crack risk value is less than or equal to the preset safety threshold, the current construction parameters are maintained; otherwise, a new construction parameter adjustment instruction is generated.

[0112] By evaluating the quality of newly solidified floor slab sections, the effectiveness of previous construction parameter adjustments can be confirmed, and changes in construction site conditions can be dynamically responded to, thereby continuously guiding the construction process and stabilizing it in an optimal state, ensuring the homogeneity and reliability of overall construction quality, and achieving continuous self-optimization control.

[0113] The preset safety threshold can be determined based on a baseline risk value. For example, it can be set to be equal to the baseline risk value, or to 1.05 to 1.10 times the baseline risk value. When the baseline risk value is 0.5, the preset safety threshold can be 0.5 or 0.525.

[0114] Please see Figure 4 As shown, a construction control system for cracks in cast-in-place floor slabs based on image processing includes: an image recognition module, used to acquire surface images of key areas of currently solidified floor slab segments and identify honeycomb defects, delamination defects, and crack defects in the surface images.

[0115] The feature calculation module is used to calculate the area ratio of cellular defects and the regional dispersion of layered defects; obtain the width and direction of crack defects, and determine the total influence based on this.

[0116] The risk assessment module is used to input the area ratio, regional dispersion, and total impact into a preset defect degree analysis model, and output the crack risk value for subsequent pouring.

[0117] The parameter adjustment module is used to generate construction parameter adjustment instructions, including pouring speed and vibration intensity, based on the crack risk value.

[0118] The instruction transmission module is used to send construction parameter adjustment instructions to the pouring equipment for pouring the next floor slab segment.

[0119] The construction feedback module is used to collect surface images of the next floor slab segment after it has solidified, determine new crack risk values, and decide whether to maintain the current construction parameters or regenerate new construction parameter adjustment instructions.

[0120] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0121] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0122] Those skilled in the art will recognize that the modules and algorithm steps of the various examples 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 implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0123] In addition, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0124] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0125] 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 within the protection scope of the present invention.

Claims

1. A method for controlling cracks in cast-in-place floor slabs based on image processing, characterized in that, include: Collect surface images of key areas of the currently solidified floor slab section and identify honeycomb defects, delamination defects, and crack defects in the surface images; Calculate the area proportion of honeycomb defects and the regional dispersion of delamination defects; obtain the width and orientation of crack defects, and determine the total influence accordingly; Input the area ratio, regional dispersion, and total impact into the preset defect degree analysis model, and output the crack risk value of subsequent pouring; Based on the crack risk value, generate construction parameter adjustment instructions including pouring speed and vibration intensity; Send the construction parameter adjustment command to the pouring equipment to pour the next floor slab segment; After the next floor slab has solidified, its surface image is acquired and a new crack risk value is determined. If the new crack risk value is less than or equal to the preset safety threshold, the current construction parameters are maintained; otherwise, a new construction parameter adjustment instruction is generated.

2. The method for controlling cracks in cast-in-place floor slabs based on image processing according to claim 1, characterized in that, The process of acquiring surface images of key areas in the currently solidified floor slab segment is as follows: The key areas are the mid-span stress zone of the currently solidified floor slab segment, the area around the beam-column joint, and the area on both sides of the construction joint. Adjust the image acquisition device to an angle where the optical axis is perpendicular to the surface of the currently solidified floor slab section; The image acquisition equipment performs full-coverage image acquisition of the mid-span stress zone, the area around the beam-column joint, and the areas on both sides of the construction joint according to the preset path scanning scheme.

3. The method for controlling cracks in cast-in-place floor slabs based on image processing according to claim 2, characterized in that, The process of establishing the path scanning scheme is as follows: Obtain the rectangular boundary dimensions of the mid-span stress zone, the annular boundary dimensions of the area surrounding the beam-column joint, and the strip boundary dimensions of the areas on both sides of the construction joint; For the mid-span stress zone, a grid-like scanning path covering the entire rectangular boundary is generated; For the area surrounding the beam-column joint, a radial scanning path is generated with the center of the beam-column as the reference, and the radial scanning path is evenly distributed in the radial and circumferential directions. For the areas on both sides of the construction joint, a bow-shaped scanning path is generated parallel to the extension direction of the construction joint, and the bow-shaped path is symmetrically distributed on both sides of the construction joint. The generated scanning path is converted into motion control commands for the image acquisition device, forming a complete path scanning scheme.

4. The method for controlling cracks in cast-in-place floor slabs based on image processing according to claim 1, characterized in that, The process of identifying honeycomb defects, delamination defects, and crack defects in surface images is as follows: The acquired surface image was segmented into multiple candidate regions, and each candidate region was then filtered as follows: Calculate the ratio of the candidate region area to the area of ​​the minimum bounding rectangle, and also calculate the change in the candidate region area before and after the morphological closing operation; If the ratio is less than the first reference benchmark and the change is greater than zero, the candidate region is identified as a cellular defect. Calculate the aspect ratio of the minimum bounding rectangle of the candidate region. If the aspect ratio is greater than the second reference benchmark, it is determined to be a strip region. Extract the gray-scale profile curve along its short side and calculate the ratio of the curve trough depth to the background gray-scale standard deviation. If the ratio is greater than the third reference benchmark, the candidate region is identified as a layered defect; The candidate region is skeletonized to obtain the central skeleton line, which is used to identify crack defects.

5. The method for controlling cracks in cast-in-place floor slabs based on image processing according to claim 4, characterized in that, The process of skeletalizing the candidate region to obtain the central skeleton line and identifying crack defects accordingly includes: Determine if the central skeleton line has a branch point; if a branch point exists, perform the following operations: Extract all branch points connecting three or more skeleton lines from the central skeleton line; Divide the central skeleton line into multiple skeleton line segments using each branch point as the boundary; The ratio of the total length of each skeleton segment to the distance between its first and last endpoints is used as the meandering degree; The maximum value among all meandering degrees is taken as the final meandering degree index for the candidate region; If the final meandering index is greater than the fourth reference benchmark, the candidate region is identified as a crack defect. Otherwise, calculate the ratio of the total length of the central skeleton line to the distance between its first and last endpoints. If the ratio is greater than the fourth reference datum, then the candidate region is identified as a crack defect.

6. The method for controlling cracks in cast-in-place floor slabs based on image processing according to claim 1, characterized in that, The calculation of the area proportion of cellular defects and the regional dispersion of layered defects is specifically as follows: The total area of ​​all cell defects is counted, and the ratio of this area to the total area of ​​critical areas is calculated to obtain the area proportion of cell defects. Obtain the geometric center coordinates of each layer of defects, and transform all geometric center coordinates to the same floor slab actual coordinate system; If there is no stratification defect, then set the regional dispersion to zero; If there is a unique layer defect, the Euclidean distance between its geometric center coordinates and the geometric center of the floor slab is used as the regional dispersion. If there are non-unique layered defects, calculate the average of the coordinates of all geometric centers in the X and Y directions to obtain the set center point of the layered defect distribution; Calculate the Euclidean distance from the coordinates of each geometric center to the center point of the set, and use the average of all Euclidean distances as the regional dispersion.

7. The method for controlling cracks in cast-in-place floor slabs based on image processing according to claim 1, characterized in that, The process of obtaining the width and orientation of the crack defect and determining the total influence accordingly is as follows: Select multiple measurement points at equal intervals along each central skeleton line, and draw a perpendicular line segment to the central skeleton line at each measurement point; For each crack defect, the average physical width of each measurement point is taken as the representative width of the crack defect. The direction of the line connecting the first and last endpoints of the central skeleton line is taken as the overall direction of the crack defect. Calculate the absolute value of the angle between the overall orientation and the reference direction, and use it as the orientation influence factor; multiply the width of the crack defect by the orientation influence factor to obtain the single-strip influence degree of the crack defect; The sum of the influence of all individual lines is taken as the total influence of the crack defect.

8. The method for controlling cracks in cast-in-place floor slabs based on image processing according to claim 1, characterized in that, The process of constructing the defect severity analysis model is as follows: Data sets were compiled by collecting the area percentage, regional dispersion, total impact, and crack risk value of solidified floor slab sections from historical construction projects. The dataset is divided into training, validation and test sets according to a set ratio. The area ratio, regional dispersion and total impact in the training set are used as input features, and the crack risk value is used as the target variable. The gradient boosting tree model is used for iterative training. The nonlinear mapping relationship between the input features and the target variable is established by minimizing the loss function, and the initial defect degree analysis model is constructed. The initial defect severity analysis model is used to predict the crack risk value of the test set. The mean square error is then used to adjust and optimize the initial defect severity analysis model, and the final defect severity analysis model is output.

9. The method for controlling cracks in cast-in-place floor slabs based on image processing according to claim 1, characterized in that, The process of generating construction parameter adjustment instructions, including pouring speed and vibration intensity, based on crack risk values ​​is as follows: Obtain the initial pouring speed and initial vibration intensity during the construction of the currently solidified floor slab segment, and use them as the benchmark pouring speed and benchmark vibration intensity; The ratio of the current crack risk value of the solidified floor slab section to the benchmark risk value is used as the risk correction coefficient. Based on the risk correction coefficient, the adjusted pouring speed is determined according to the first functional relationship, and the adjusted vibration intensity is determined according to the second functional relationship; wherein, the first functional relationship and the second functional relationship are piecewise functional relationships, and the adjusted pouring speed decreases monotonically with the increase of the risk correction coefficient, and the adjusted vibration intensity increases monotonically with the increase of the risk correction coefficient. The adjusted pouring speed is limited between the preset lower and upper speed thresholds to obtain the final pouring speed; the vibration strength is handled in the same way. Generate construction parameter adjustment instructions that include the final pouring speed and the final vibration intensity.

10. A construction control system for cracks in cast-in-place floor slabs based on image processing, characterized in that, include: The image recognition module is used to acquire surface images of key areas of the currently solidified floor slab section and identify honeycomb defects, delamination defects, and crack defects in the surface images; The feature calculation module is used to calculate the area ratio of cellular defects and the regional dispersion of layered defects; obtain the width and orientation of crack defects, and determine the total influence based on this. The risk assessment module is used to input the area ratio, regional dispersion and total impact into the preset defect degree analysis model, and output the crack risk value of subsequent pouring. The parameter adjustment module is used to generate construction parameter adjustment instructions, including pouring speed and vibration intensity, based on the crack risk value. The instruction transmission module is used to send construction parameter adjustment instructions to the pouring equipment for pouring the next floor slab segment; The construction feedback module is used to collect surface images of the next floor slab segment after it has solidified, determine new crack risk values, and decide whether to maintain the current construction parameters or regenerate new construction parameter adjustment instructions.

Images