Method and device for determining pouring parameters for precast beam concrete defect, equipment and medium

Abstract

This invention provides a method, apparatus, equipment, and medium for determining pouring parameters for concrete defects in precast beams. The method for determining pouring parameters for concrete defects in precast beams includes: acquiring an image of the precast beam and obtaining the pixel regions of concrete surface defects and the confidence levels corresponding to each defect category through an instance segmentation model; determining the physical area of ​​the concrete surface defects based on the mapping relationship between the pixel regions and the actual dimensions of the precast beam; using the confidence levels of each defect category as the defect category membership degree, and determining the pouring effect membership degree based on the pouring effect evaluation value of the previous pouring cycle; determining the pouring coefficient within the current pouring cycle through fuzzy inference based on the defect category membership degree and the pouring effect membership degree; determining the concrete weight parameters used for pouring the concrete surface defects based on the physical area and the pouring coefficient; and acquiring defect images before and after the current pouring to determine the pouring effect evaluation value for the current pouring cycle.

Description

Technical Field

[0001] This invention relates to the field of image processing and fuzzy control technology, and in particular to a method, apparatus, equipment and medium for determining pouring parameters for concrete defects in precast beams. Background Technology

[0002] With the development of image recognition technology, various image recognition-based methods for detecting defects on concrete surfaces are gradually being applied to the quality inspection of precast components. These technologies typically acquire images of the precast beam surface, use image processing models to identify defect areas and determine defect types, thereby assisting in subsequent quality control.

[0003] Existing technologies generally employ fixed rules or empirical parameters to directly set the corresponding pouring or repair volume based on the defect category. However, since the morphology, severity, and repair effect of defects vary with actual conditions, fixed pouring strategies are often difficult to adapt to the differences between different defects, and cannot dynamically optimize the pouring strategy based on feedback after repair. This results in inaccurate pouring parameters and unstable repair effects. Summary of the Invention

[0004] This invention provides a method, apparatus, equipment, and medium for determining pouring parameters for precast beam concrete defects, in order to solve the problem in the prior art that it is difficult to dynamically adjust pouring parameters after identifying surface defects in precast beams.

[0005] This invention provides a method for determining pouring parameters for concrete defects in precast beams, comprising: acquiring an image of the precast beam and obtaining pixel regions of concrete surface defects and confidence levels corresponding to each defect category through an instance segmentation model; determining the physical area of ​​the concrete surface defects based on the mapping relationship between the pixel regions and the actual dimensions of the precast beam; using the confidence level of each defect category as the defect category membership degree, and determining the pouring effect membership degree based on the pouring effect evaluation value of the previous pouring cycle; determining the pouring coefficient within the current pouring cycle through fuzzy inference based on the defect category membership degree and the pouring effect membership degree; determining the concrete weight parameters for pouring the concrete surface defects based on the physical area and the pouring coefficient; and acquiring defect images before and after the current pouring to determine the pouring effect evaluation value for the current pouring cycle.

[0006] According to the method for determining pouring parameters for precast beam concrete defects provided by the present invention, the step of determining the pouring coefficient within the current pouring cycle through fuzzy reasoning based on the membership degree of the defect category and the membership degree of the pouring effect includes: matching preset fuzzy rules according to the membership degree of the defect category and the membership degree of the pouring effect to determine the triggered pouring strategy and its corresponding activation intensity; trimming the membership function corresponding to each pouring strategy according to the activation intensity to obtain multiple trimmed output fuzzy sets; merging the output fuzzy sets to form a comprehensive fuzzy set; defuzzifying the comprehensive fuzzy set according to the shape centroid method to determine the pouring coefficient adjustment amount; and adjusting the pouring coefficient of the previous pouring cycle according to the pouring coefficient adjustment amount to obtain the pouring coefficient of the current pouring cycle.

[0007] According to the method for determining concrete defect pouring parameters for precast beams provided by the present invention, the step of acquiring defect images before and after the current pouring to determine the pouring effect evaluation value of the current pouring cycle includes: acquiring a first defect image of the concrete surface defects before the current pouring and a second defect image after the current pouring; determining the first defect area before the current pouring and the first confidence level corresponding to each defect type based on the first defect image; determining the second defect area after the current pouring and the second confidence level corresponding to each defect type based on the second defect image; and calculating the pouring effect evaluation value of the current pouring cycle based on the first defect area, the first confidence level, the second defect area, the second confidence level, and a preset defect severity weight vector.

[0008] According to the method for determining the pouring parameters of concrete defects in precast beams provided by the present invention, the method further includes: converting the pixel coordinates in the pixel region into the local actual coordinates of the concrete surface defects based on the mapping relationship between the pixel size of the pixel region and the actual size of the precast beam; establishing a global coordinate system for the precast beam with a preset point as the origin, and translating the local actual coordinates into the global coordinate system to determine the physical location of the concrete surface defects on the precast beam.

[0009] According to the method for determining pouring parameters for precast beam concrete defects provided by the present invention, the step of determining the concrete weight parameters for pouring the concrete surface defects based on the physical area and pouring coefficient includes: dividing the concrete surface defects into multiple areas to be poured; determining the center position and area of ​​each area to be poured; determining the pouring position based on the center position of each area to be poured; and determining the concrete weight parameters of each area to be poured based on the area of ​​each area to be poured.

[0010] According to the method for determining the pouring parameters of precast beam concrete defects provided by the present invention, dividing the concrete surface defects into multiple pouring areas includes: determining the circumscribed rectangle of the concrete surface defects based on the pixel area, and dividing the circumscribed rectangle equally in the horizontal and vertical directions to form multiple pouring areas.

[0011] The method for determining concrete defect casting parameters for precast beams according to the present invention further includes updating the instance segmentation model based on newly acquired defect images to improve defect identification accuracy.

[0012] This invention also provides a device for determining pouring parameters for concrete defects in precast beams, comprising: a segmentation module for acquiring images of the precast beam and obtaining pixel regions of concrete surface defects and confidence levels corresponding to each defect category through an instance segmentation model; an area determination module for determining the physical area of ​​the concrete surface defects based on the mapping relationship between the pixel regions and the actual dimensions of the precast beam; a membership determination module for using the confidence levels of each defect category as the defect category membership degree and determining the pouring effect membership degree based on the pouring effect evaluation value of the previous pouring cycle; a pouring coefficient determination module for determining the pouring coefficient within the current pouring cycle through fuzzy reasoning based on the defect category membership degree and the pouring effect membership degree; a pouring quantity parameter determination module for determining the concrete weight parameter used for pouring the concrete surface defects based on the physical area and the pouring coefficient; and an evaluation module for acquiring defect images before and after the current pouring to determine the pouring effect evaluation value for the current pouring cycle.

[0013] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for determining the casting parameters for precast beam concrete defects as described above.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for determining the casting parameters for precast beam concrete defects as described above.

[0015] The present invention provides a method, apparatus, equipment, and medium for determining pouring parameters for precast beam concrete defects. It obtains pixel regions of concrete surface defects through an instance segmentation model and uses the identified defect category confidence level as the defect category membership degree. Simultaneously, it uses the pouring effect evaluation value from the previous pouring cycle to determine the pouring effect membership degree. By combining these two membership degrees through fuzzy inference, the pouring coefficient can be dynamically determined in different pouring cycles. This allows the pouring parameters to be adaptively adjusted according to changes in defect type and actual pouring effect. Therefore, the accuracy and stability of pouring can be gradually improved through dynamic adjustment, thereby enhancing the effectiveness and efficiency of concrete defect repair. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of a fabric distribution machine according to an embodiment of the present invention;

[0018] Figure 2 This is a schematic diagram of precast beam images acquired by an industrial camera according to an embodiment of the present invention;

[0019] Figure 3 This is a flowchart of a method for determining concrete defect casting parameters for precast beams according to an embodiment of the present invention;

[0020] Figure 4 This is a flowchart of determining the pouring coefficient within the current pouring cycle based on the defect category membership degree and the pouring effect membership degree according to an embodiment of the present invention;

[0021] Figure 5 This is a schematic diagram illustrating the merging of various output fuzzy sets to form a comprehensive fuzzy set according to an embodiment of the present invention;

[0022] Figure 6 This is a flowchart of obtaining defect images before and after the current pouring according to an embodiment of the present invention, in order to determine the evaluation value of the pouring effect in this pouring cycle;

[0023] Figure 7 This is a flowchart for determining the concrete weight parameters used to pour concrete for surface defects based on physical area and pouring coefficient.

[0024] Figure 8 This diagram shows a structural block diagram of a device for determining concrete defect casting parameters for precast beams according to an embodiment of the present invention.

[0025] Figure 9 This is a structural block diagram of an electronic device suitable for determining the pouring parameters of precast beam concrete defects according to an embodiment of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0027] In related technologies, after identifying casting defects on the surface of precast beams through image recognition, casting parameters are often determined only by fixed rules or empirical parameters. It is difficult to dynamically adjust the casting strategy according to changes in the actual situation, resulting in unstable repair effects on defects.

[0028] In view of this, the present invention provides a method for determining the pouring parameters for concrete defects in precast beams.

[0029] Figure 1 This is a schematic diagram of a fabric feeding machine according to an embodiment of the present invention.

[0030] The replenishment of precast beam 200 often requires the use of a concrete placing machine 100 that moves along the length of the precast beam 200, such as... Figure 1 As shown, the concrete placing boom 100 is generally equipped with two hanging columns 110 and a crossbeam 120 on one side. The concrete placing boom 100 also has a discharge port 130 for feeding the precast beam 200. A motor and hydraulic pump are placed above the crossbeam 120 to provide power to the piston rod, driving the movement of the concrete placing boom. To acquire images of the precast beam surface, an industrial camera 300 can be mounted on a camera bracket 310, which is then mounted on the concrete placing boom 100. Furthermore, to ensure the quality of image acquisition, two strip light sources 140 are installed on both sides of the crossbeam 120 to ensure consistent ambient lighting.

[0031] Given the above setup, the industrial camera has a relatively short shooting distance and a large field of view, so it is necessary to determine the relevant parameters for image acquisition.

[0032] Specifically, the vertical distance between the industrial camera lens and the surface of the precast beam is set as the shooting height. The field of view formed by the industrial camera on the surface of the precast beam is set to... × ,in, Let be the actual lateral coverage length of a single camera field of view on the beam surface, and be the length of the long side of the camera's field of view. Let be the actual longitudinal coverage width of a single camera field of view on the beam surface, and be the short side dimension of the camera's field of view; set the long side of the sensor as . The shorter side is Lens focal length The formula for calculating sensor size selection is as follows:

[0033]

[0034] Taking a specific case as an example, assuming the working distance of the industrial camera is 1.5m, and the longer side of the field of view... It can cover a precast beam width of 2.85m. Substituting into the above formula, we get:

[0035]

[0036] To avoid image distortion, the focal length should be as large as possible. Considering that both focal length and sensor size have fixed and commonly used dimensions, a 6mm focal length lens is selected here. The long side of the sensor is 11.4mm, which is closer to the size of a 1-inch (12.8mm × 9.6mm) sensor. In this case, a 6mm focal length camera lens and a 1-inch camera sensor are selected.

[0037] Figure 2 This is a schematic diagram of precast beam images acquired by an industrial camera according to an embodiment of the present invention.

[0038] like Figure 2 As shown, due to the long length of the precast beam, multi-point imaging can be used to acquire images of its surface. During image acquisition, the industrial camera moves to a designated position to capture images. Let the physical length of the long side of the precast beam be... The physical length of the short side of the precast beam is The field of view size is × The longer side of the field of view Slightly larger than the short side length of the precast beam In this embodiment, the industrial camera is fixed to the crossbeam of the fabric laying machine, allowing the optical axis to move along the centerline of the precast beam. The first image captured precisely covers the top left side of the precast beam. The position of the Nth image is N×b away from the first image, and the total number of images captured is A / b rounded up.

[0039] According to embodiments of the present invention, image acquisition can be performed using an overlapping imaging method. By controlling adjacent images to maintain a predetermined overlapping area, for example, the overlap rate between adjacent images can be 16.7%, achieving seamless coverage of the precast beam surface. The acquired adjacent images are registered using the ORB feature point matching algorithm, stitching all acquired images together to form a single image of the precast beam's casting. To improve the robustness of the matching, a Random Sample Consensus Algorithm (RANSAC) is introduced to eliminate mismatched points (outside points), calculating the homography matrix only based on correct matching points (inside points), thereby accurately estimating the transformation relationship between images and achieving robust image stitching. The stitched image is cropped according to a width of b to facilitate input into the instance segmentation model for detection.

[0040] Figure 3 This is a flowchart of a method for determining the pouring parameters for precast beam concrete defects according to an embodiment of the present invention.

[0041] like Figure 3 As shown, the method for determining the pouring parameters for concrete defects in precast beams includes steps S310 to S360.

[0042] In step S310, an image of the precast beam is acquired, and the pixel regions of concrete surface defects and the confidence levels corresponding to each defect category are obtained through the instance segmentation model.

[0043] According to embodiments of the present invention, the instance segmentation model can be any existing model with instance segmentation capabilities, such as Mask R-CNN, YOLO-Seg, or Detectron2, and the present invention does not limit this to any particular model. After processing by the instance segmentation model, a mask for labeling concrete surface defects can be output, along with a category label representing the defect category and the confidence level of the probability that the defect belongs to each defect category.

[0044] For example, based on the size of the concrete surface defect, the defect can be classified into small defects, medium defects, and large defects.

[0045] In step S320, the physical area of ​​the concrete surface defects is determined based on the mapping relationship between the pixel region and the actual size of the precast beam.

[0046] According to an embodiment of the present invention, since there is a linear mapping relationship between the pixel size of the image containing the pixel region and the actual size of the precast beam, the pixel region can be transformed into the world coordinate system, thereby determining the physical area of ​​the concrete surface defect.

[0047] Specifically, let the pixel size of the image including the pixel region be... Where c is the longer side of the image, d is the shorter side of the image, and N is the number of defective pixels in the pixel region. maskThe area ratio of the pixel region is:

[0048]

[0049] Based on the linear mapping relationship, the actual physical area of ​​concrete surface defects is:

[0050]

[0051] In step S330, the confidence level of each defect category is used as the defect category membership degree, and the membership degree of the pouring effect is determined based on the pouring effect evaluation value of the previous pouring cycle.

[0052] According to an embodiment of the present invention, the defect categories are divided into large defects, medium defects, and small defects according to their area size, and the confidence level of the concrete surface defects corresponding to each defect category is given according to the instance segmentation model, and the confidence level is used as the defect category membership degree.

[0053] According to an embodiment of the present invention, the evaluation value of the pouring effect of the previous pouring cycle is calculated by comparing the changes in the defect area and confidence level of the images before and after the previous pouring.

[0054] According to an embodiment of the present invention, the evaluation value of the pouring effect in the previous pouring cycle can be converted into the pouring effect membership degree according to a preset trapezoidal membership function.

[0055] According to an embodiment of the present invention, the pouring result can be defined as three types: poor pouring effect (P), average pouring effect (F), and good pouring effect (G).

[0056] For example, the trapezoidal membership functions of the above three sets are as follows:

[0057] Poor re-watering effect (P): Completely belongs to: E<=0.3 (μ=1), Completely does not belong to: E>=0.5 (μ=0), Linear decreasing interval: (0.3, 0.5), Membership function μ under the condition of poor re-watering effect. poor (E) is: μ poor (E) = (0.5 - E) / (0.5 - 0.3).

[0058] The re-watering effect is mediocre (F): Completely belongs to: 0.4 <= E <= 0.6 (μ = 1), linearly increasing interval: (0.2, 0.4), membership function μ under the condition of mediocre re-watering effect. fair (E) is: μ fair (E) = (E - 0.2) / (0.4 - 0.2), linear decreasing interval: (0.6, 0.8), formula: μ fair (E) = (0.8 - E) / (0.8 - 0.6).

[0059] Good re-watering effect (G): Completely belongs to: E>=0.7 (μ=1), Completely does not belong to: E<=0.5 (μ=0), Linearly increasing interval: (0.5, 0.7), Membership function μ under the condition of good re-watering effect. good (E) is: μ good (E) = (E - 0.5) / (0.7 - 0.5). (For example, when E = 5.5, substituting into the trapezoidal membership function, the fuzzy membership is [poor repair effect: 0.0, average repair effect: 1.0, good repair effect: 0.25]).

[0060] Where E is the evaluation value of the pouring effect, and μ is the membership degree of the pouring effect.

[0061] In step S340, the pouring coefficient within the current pouring cycle is determined by fuzzy reasoning based on the membership degree of the defect category and the membership degree of the pouring effect.

[0062] According to an embodiment of the present invention, a preset fuzzy rule base is matched to determine the triggered pouring strategy based on the membership degree of the defect category and the membership degree of the pouring effect, and the corresponding activation intensity is obtained. The output membership function of each pouring strategy is pruned according to the activation intensity, and the pruned output fuzzy sets are merged. The shape centroid method is used to defuzzify the output fuzzy set to obtain the pouring coefficient adjustment ΔK of the current pouring cycle relative to the previous pouring cycle. ΔK is applied to the pouring coefficient of the previous cycle to obtain the pouring coefficient K of the current pouring cycle.

[0063] In an embodiment of the present invention, the initial pouring coefficient of the first pouring cycle can be determined through a prior calibration test. Specifically, samples with three typical defect types—small, medium, and large defects—can be selected, and their defect areas can be actually measured to obtain the corresponding physical areas. Subsequently, a concrete placing boom was used to repair the aforementioned defects, and a weighing device recorded the weight of concrete consumed during the repair. The initial pouring coefficients for the three types of defects can be calculated separately:

[0064]

[0065] in , and These correspond to the initial pouring coefficients for small, medium, and large defects, respectively.

[0066] During the subsequent pouring process, the pouring coefficient can be iterated based on the initial pouring coefficient to meet the actual pouring requirements.

[0067] In step S350, the concrete weight parameters for pouring concrete surface defects are determined based on the physical area and the pouring coefficient.

[0068] According to an embodiment of the present invention, the concrete weight parameter t used for pouring the concrete surface defect is determined based on the physical area S of the defect and the pouring coefficient K, specifically as follows:

[0069]

[0070] In step S360, defect images before and after this pouring are obtained to determine the evaluation value of the pouring effect in this pouring cycle.

[0071] According to an embodiment of the present invention, images of concrete surface defects are collected before and after pouring to determine the evaluation value of the pouring effect in this pouring cycle, providing a basis for adjusting the pouring coefficient in the next cycle.

[0072] By incorporating the confidence level of the defect category and the evaluation value of the pouring effect into the fuzzy inference, the pouring coefficient can be dynamically adjusted between different pouring cycles, and the corresponding concrete weight parameters can be calculated based on the physical area of ​​the defect. This enables the determination of dynamic pouring parameters based on visual recognition results, ensuring the pouring effect.

[0073] Figure 4 This is a flowchart illustrating how, according to an embodiment of the present invention, the pouring coefficient within a pouring cycle is determined through fuzzy reasoning based on the membership degree of the defect category and the membership degree of the pouring effect.

[0074] like Figure 4 As shown, step S340 includes steps S410 to S450.

[0075] In step S410, a preset fuzzy rule is matched based on the membership degree of the defect category and the membership degree of the pouring effect to determine the triggered pouring strategy and its corresponding activation intensity.

[0076] According to an embodiment of the present invention, the preset fuzzy rule is constructed in the form of "if (precondition), then (output conclusion)", wherein the precondition is jointly determined by the membership degree of the defect category and the membership degree of the pouring effect, and the output conclusion is the adjustment strategy of the corresponding pouring coefficient.

[0077] In fuzzy rules, the output of each rule points to a certain type of casting coefficient (K). S K M or K L These correspond to the pouring coefficients for small, medium, and large defects, respectively, and their adjustment directions. The fuzzy linguistic variables for the adjustment direction include: significantly reduced (LD), slightly reduced (SD), unchanged (NC), slightly increased (SI), and significantly increased (LI), and their corresponding adjustment amounts ΔK are defined within a preset universe of discourse.

[0078] The activation strength of each rule is determined by the minimum value of the membership degrees of its premises, which follows the "AND" operation principle in fuzzy logic.

[0079] In step S420, the membership functions corresponding to each pouring strategy are clipped according to the activation intensity to obtain multiple clipped output fuzzy sets.

[0080] According to an embodiment of the present invention, after triggering the corresponding fuzzy rules based on the membership degree of the defect category and the membership degree of the casting effect, each triggered rule has a corresponding activation strength. To reflect the contribution of different rules in this inference, the output membership function corresponding to each rule needs to be pruned.

[0081] Specifically, for a given activated rule, its output corresponds to a membership function representing a specific direction of casting coefficient adjustment (e.g., slightly increasing SI, keeping NC unchanged, etc.). Let the activation strength of this rule be... Then, the membership function of the adjustment direction is truncated at the top (pruned), and its original membership function is removed. Limited to no more than Within the range, thus obtaining the cropped output fuzzy set:

[0082]

[0083] in, The membership degree after clipping. The domain of discussion for adjusting the pouring coefficient is defined.

[0084] In step S430, the output fuzzy sets are merged to form a comprehensive fuzzy set.

[0085] According to an embodiment of the present invention, after obtaining the trimmed output fuzzy sets corresponding to each triggered rule, these output fuzzy sets need to be merged to form a comprehensive fuzzy set for fuzzy inference. Specifically, the "maximum operation" in fuzzy logic can be used, that is, the maximum value is taken for the membership degree of all trimmed output fuzzy sets on the same domain, so that each rule has a comprehensive impact on the final pouring coefficient adjustment.

[0086] In step S440, the comprehensive fuzzy set is defuzzified according to the shape centroid method to determine the adjustment amount of the casting coefficient.

[0087] According to an embodiment of the present invention, after obtaining the comprehensive fuzzy set, a defuzzification method is needed to determine the precise adjustment amount of the casting coefficient. This embodiment uses the shape centroid method (also known as the centroid method), which is commonly used in fuzzy control, to defuzzify the comprehensive fuzzy set.

[0088] Specifically, let the universe of discourse for the adjustment of the pouring coefficient be... The membership function of the comprehensive fuzzy set on this universe is: According to the center of gravity method, the adjustment amount of the pouring coefficient. It can be obtained using the following formula:

[0089]

[0090] In this equation, the numerator represents the "torque" of the comprehensive fuzzy set in the universe of discourse, the denominator represents the cumulative amount of comprehensive membership, and the ratio of these values ​​is the centroid position of the comprehensive fuzzy set. The abscissa of the centroid is then used as the precise adjustment amount for the casting coefficient.

[0091] The universe of discourse can be discretized into several equally spaced sampling points. And calculate the comprehensive membership degree at each sampling point. The adjustment amount can be calculated using the following discretization formula:

[0092]

[0093] Through the above defuzzification process, the adjustment amount of the pouring coefficient for the current pouring cycle relative to the previous pouring cycle can be obtained from the comprehensive fuzzy set.

[0094] In step S450, the pouring coefficient of the previous pouring cycle is adjusted according to the pouring coefficient adjustment amount to obtain the pouring coefficient of the current pouring cycle.

[0095] Figure 5 This is a schematic diagram illustrating the merging of various output fuzzy sets to form a comprehensive fuzzy set according to an embodiment of the present invention.

[0096] like Figure 5 As shown in the example, the preset fuzzy rules are shown in the table below:

[0097] Table 1 shows the fuzzy rules for adjusting the casting coefficient according to an embodiment of the present invention.

[0098]

[0099] Assuming the defect type membership degrees are: [Small defect: 0.25, Medium defect: 0.60, Large defect: 0.15], and the pouring effect membership degrees are: [Poor pouring effect: 0.0, Average pouring effect: 1.0, Good pouring effect: 0.25]. Since the defect type with the highest membership degree is "defect," only the pouring coefficient K needs to be used during pouring. M The amount of additional water needed is calculated, so only K is considered. M After adjustments, rules 5 and 6 are satisfied. The activation intensity of rule 5 is min(0.60,1)=0.60, and the activation intensity of rule 6 is min(0.60,0.25)=0.25.

[0100] The theoretical range of the change in the casting coefficient ΔK is defined as [-0.3, +0.3], representing a maximum adjustment range of ±30%. ΔK is divided into the five fuzzy subsets mentioned above: {significant decrease (LD), slight decrease (SD), unchanged (NC), slight increase (SI), significant increase (LI)}. Each subset uses a triangular membership function, the parameters of which are defined in the table below:

[0101] Table 2 lists the parameters of the triangular membership function for each fuzzy subset according to an embodiment of the present invention.

[0102]

[0103] Furthermore, the top of the triangular membership function is pruned according to the activation strength of different rules. Then, the largest of all pruned output fuzzy sets is merged to finally form K. M The independent integrated output fuzzy set.

[0104] In the example above, Rule 5 slightly increases the activation strength to 0.60, while Rule 6 keeps the activation strength unchanged at 0.25. This forms a triangular function with boundaries (0, 0.2) and a vertex at 0.1, and a triangular function with boundaries (-0.1, +0.1) and a vertex at 0. This triangular function is the comprehensive fuzzy set.

[0105] Further solve for the pouring coefficient adjustment ΔK of this comprehensive fuzzy set. M In this embodiment, both the universe of discourse and membership degree are discretized with a step size of 0.05 and substituted into the discretization formula described above, finally yielding the casting coefficient adjustment amount ΔK. M =0.1225 / 2.35≈0.0521, which means we can use K from the previous pouring cycle. M The adjustment was made upwards by 5.21% based on the previous figure.

[0106] Figure 6 This is a flowchart based on an embodiment of the present invention for obtaining defect images before and after the current pouring to determine the evaluation value of the pouring effect in this pouring cycle.

[0107] like Figure 6 As shown, step S360 includes steps S610 to S640.

[0108] In step S610, a first defect image of the concrete surface before this pouring and a second defect image after this pouring are obtained.

[0109] In step S620, the area of ​​the first defect before this pouring and the first confidence level corresponding to each defect type are determined based on the first defect image.

[0110] In step S630, the area of ​​the second defect after this pouring and the second confidence level corresponding to each defect type are determined based on the second defect image.

[0111] In step S640, the evaluation value of the pouring effect for this pouring cycle is calculated based on the first defect area, the first confidence level, the second defect area, the second confidence level, and the preset defect severity weight vector.

[0112] According to an embodiment of the present invention, the evaluation value E of the pouring effect can be calculated using the following formula:

[0113]

[0114] Where S before S is the area of ​​the first defect. after This represents the area of ​​the second defect. This is the first defect confidence vector, composed of the first confidence levels corresponding to each defect type. , as well as These represent the confidence levels for classifying defects as "small defects," "medium defects," and "large defects" before pouring; Let be the second defect confidence vector composed of the second confidence levels corresponding to each defect type, where , as well as These represent the confidence levels for classifying defects as "small defects," "medium defects," and "large defects" after pouring. This is the defect severity weight vector. , as well as The severity weights are respectively for "minor shortage", "medium shortage" and "major shortage".

[0115] In one illustrative embodiment, the method for determining the pouring parameters for precast beam concrete defects further includes:

[0116] Based on the mapping relationship between the pixel size of the pixel region and the actual size of the precast beam, the pixel coordinates in the pixel region are converted into the local actual coordinates of the defects on the concrete surface.

[0117] A global coordinate system for the precast beam is established with the preset points of the precast beam as the origin. The local actual coordinates are translated into the global coordinate system to determine the physical location of concrete surface defects on the precast beam.

[0118] According to an embodiment of the present invention, let c be the pixel size of the image along the longitudinal direction, and d be the pixel size of the image along the transverse direction. The actual dimensions of the precast beam in the transverse and longitudinal directions are respectively... and For any pixel within a pixel region, its pixel coordinates can be mapped to its local actual coordinates through a linear scaling transformation, i.e.:

[0119]

[0120] in, These are pixel coordinates in the image coordinate system. This represents the local actual coordinates of the pixel coordinates mapped onto the surface of the precast beam. A global coordinate system is established with a fixed point on the precast beam (such as the beam end or construction reference point) as the origin. By shifting the aforementioned local actual coordinates to the global coordinate system by an offset, the global physical location of the defect area on the precast beam can be obtained.

[0121]

[0122] in, This represents the reference offset of the corresponding region in the image within the global coordinate system of the precast beam.

[0123] By following the steps described above, the physical location of concrete surface defects on the precast beam can be determined.

[0124] Figure 7 This is a flowchart for determining the concrete weight parameters used to pour concrete for surface defects based on physical area and pouring coefficient.

[0125] like Figure 7 As shown, step S350 includes steps S710 to S730.

[0126] In step S710, the concrete surface defects are divided into multiple areas to be poured.

[0127] According to an embodiment of the present invention, the circumscribed rectangle of the defect can be determined based on the pixel region, and this circumscribed rectangle can be used as the division range. Specifically, the circumscribed rectangle can be divided equally in both the horizontal and vertical directions to form multiple areas to be poured with basically the same size.

[0128] In step S720, the center position and area of ​​each area to be poured are determined.

[0129] According to an embodiment of the present invention, after the areas to be poured are divided, the center position of each area can be calculated, which can be determined by the geometric center of the boundary of the area to be poured.

[0130] Schematic, the pixel coordinates of the geometric center in the image coordinate system can be represented as:

[0131]

[0132] in, and The pixel coordinates of the geometric center and These are the pixel x-coordinates of the left and right boundaries of the area to be poured, respectively. and These are the pixel coordinates of the upper and lower boundaries of the area to be poured, respectively. Then, based on the mapping relationship between the pixel coordinates of the geometric center and the actual dimensions of the precast beam, they can be converted into the physical coordinates of the pouring location in actual space.

[0133] It should be noted that the determination of the center position and area mentioned above is not limited to specific formulas, and can be achieved by geometric center, centroid estimation or area measurement based on image processing.

[0134] In step S730, the center position of each area to be poured is determined, and the concrete weight parameters of each area to be poured are determined according to the area of ​​each area to be poured.

[0135] According to an embodiment of the present invention, the concrete weight parameters corresponding to the previously determined concrete surface defects can be used to determine the concrete weight parameters of each area to be poured based on the area ratio of each area to be poured.

[0136] With the above setup, because the concrete surface defects are irregular in shape and vary in size, it would be difficult and inefficient to have the concrete placing machine place the concrete exactly according to the shape of the defect. If the center of the defect is filled with concrete at a single point, it is difficult to ensure complete filling of the defect area when the defect area is large, considering the rheological properties of concrete. Dividing the concrete surface defect into multiple areas to be poured, and then pouring in small amounts multiple times, can ensure the pouring effect.

[0137] In one illustrative embodiment, the concrete surface defect is divided into multiple areas to be poured, including: determining the circumscribed rectangle of the concrete surface defect based on the pixel region, and dividing the circumscribed rectangle into equal intervals in the horizontal and vertical directions to form multiple areas to be poured.

[0138] In one illustrative embodiment, the method for determining the pouring parameters of precast beam concrete defects further includes updating the instance segmentation model based on newly acquired defect images to improve defect identification accuracy.

[0139] According to an embodiment of the present invention, defect images acquired during daily production can be stored in a preset directory and categorized according to shooting time, production line number, or task number, serving as a data source for subsequent model updates. Concrete surface defects are then labeled using a labeling tool, and the labeled image data is added to the original training set to ensure sample diversity. The expanded training set is then iteratively trained using the original training architecture to optimize the instance segmentation model.

[0140] Based on the above-described method for determining the pouring parameters of precast beam concrete defects, this invention also provides a device for determining the pouring parameters of precast beam concrete defects. The following is in conjunction with... Figure 8 The device is described in detail.

[0141] Figure 8 The diagram shows a structural block diagram of a device for determining the pouring parameters of precast beam concrete defects according to an embodiment of the present invention.

[0142] like Figure 8 As shown, the precast beam concrete defect pouring parameter determination device 800 of this embodiment includes a segmentation module 810, an area determination module 820, a membership degree determination module 830, a pouring coefficient determination module 840, a pouring volume parameter determination module 850, and an evaluation module 860.

[0143] The segmentation module 810 is used to acquire images of the precast beam and obtain pixel regions of concrete surface defects and the confidence levels corresponding to each defect category through an instance segmentation model. In one embodiment, the segmentation module 810 can be used to perform step S310 described above, which will not be repeated here.

[0144] The area determination module 820 is used to determine the physical area of ​​concrete surface defects based on the mapping relationship between pixel regions and the actual dimensions of the precast beam. In one embodiment, the area determination module 820 can be used to perform step S320 described above, which will not be repeated here.

[0145] The membership determination module 830 is used to use the confidence level of each defect category as the membership level of the defect category, and to determine the membership level of the pouring effect based on the pouring effect evaluation value of the previous pouring cycle. In one embodiment, the membership determination module 830 can be used to perform the step S330 described above, which will not be repeated here.

[0146] The pouring coefficient determination module 840 is used to determine the pouring coefficient within the current pouring cycle through fuzzy reasoning based on the defect category membership degree and the pouring effect membership degree. In one embodiment, the pouring coefficient determination module 840 can be used to execute the step S340 described above, which will not be repeated here.

[0147] The pouring volume parameter determination module 850 is used to determine the concrete weight parameters for pouring concrete to fill surface defects based on the physical area and pouring coefficient. In one embodiment, the pouring volume parameter determination module 850 can be used to perform step S350 described above, which will not be repeated here.

[0148] The evaluation module 860 is used to acquire defect images before and after the current pouring to determine the evaluation value of the pouring effect for this pouring cycle. In one embodiment, the evaluation module 860 can be used to perform step S360 described above, which will not be repeated here.

[0149] According to an embodiment of the present invention, the casting coefficient determination module 840 includes an activation intensity determination submodule, a fuzzy set output submodule, a merging submodule, a casting coefficient adjustment amount determination submodule, and a casting coefficient determination submodule.

[0150] The activation intensity determination submodule is used to match preset fuzzy rules based on the membership degree of defect category and pouring effect to determine the triggered pouring strategy and its corresponding activation intensity.

[0151] The fuzzy set output submodule is used to trim the membership function corresponding to each pouring strategy according to the activation intensity, and obtain multiple trimmed output fuzzy sets.

[0152] The merging submodule is used to merge the output fuzzy sets to form a comprehensive fuzzy set.

[0153] The submodule for determining the adjustment amount of the pouring coefficient is used to defuzzify the comprehensive fuzzy set according to the shape centroid method in order to determine the adjustment amount of the pouring coefficient.

[0154] The pouring coefficient determination submodule is used to adjust the pouring coefficient of the previous pouring cycle according to the pouring coefficient adjustment amount, so as to obtain the pouring coefficient of the current pouring cycle.

[0155] According to an embodiment of the present invention, the evaluation module 860 includes a defect image acquisition submodule, a first defect parameter determination submodule, a second defect parameter determination submodule, and a casting effect evaluation value determination submodule.

[0156] The defect image acquisition submodule is used to acquire the first defect image of the concrete surface before this pouring, and the second defect image after this pouring.

[0157] The first defect parameter determination submodule is used to determine the area of ​​the first defect before this pouring, and the first confidence level corresponding to each defect type, based on the first defect image.

[0158] The second defect parameter determination submodule is used to determine the area of ​​the second defect after this pouring, as well as the second confidence level corresponding to each defect type, based on the second defect image.

[0159] The submodule for determining the evaluation value of pouring effect is used to calculate the evaluation value of the pouring effect for this pouring cycle based on the first defect area, the first confidence level, the second defect area, the second confidence level, and the preset defect severity weight vector.

[0160] According to an embodiment of the present invention, the device 800 for determining the pouring parameters of precast beam concrete defects further includes a coordinate transformation module and a physical location determination module.

[0161] The coordinate transformation module is used to convert the pixel coordinates in the pixel area into the local actual coordinates of the concrete surface defects based on the mapping relationship between the pixel size of the pixel area and the actual size of the precast beam.

[0162] The physical location determination module is used to establish a global coordinate system for the precast beam with the preset points of the precast beam as the origin, and to translate the local actual coordinates to the global coordinate system in order to determine the physical location of concrete surface defects on the precast beam.

[0163] According to an embodiment of the present invention, the pouring volume parameter determination module 850 includes a region division submodule, a region to be poured parameter determination submodule, and a pouring control parameter submodule.

[0164] The area division submodule is used to divide concrete surface defects into multiple areas to be poured.

[0165] The submodule for determining parameters of the area to be poured is used to determine the center position and area of ​​each area to be poured.

[0166] The pouring control parameter submodule is used to determine the pouring position by identifying the center position of each area to be poured, and to determine the concrete weight parameters of each area to be poured based on its area.

[0167] According to an embodiment of the present invention, the region division submodule further includes an equidistant division unit.

[0168] The equidistant division unit is used to determine the outer rectangle of the concrete surface defect based on the pixel region, and to divide the outer rectangle into equidistant sections in the horizontal and vertical directions to form multiple areas to be poured.

[0169] According to an embodiment of the present invention, the device 800 for determining the pouring parameters of precast beam concrete defects further includes a model update module.

[0170] The model update module is used to update the instance segmentation model based on newly acquired defect images to improve defect recognition accuracy.

[0171] According to embodiments of the present invention, any plurality of modules among the segmentation module 810, area determination module 820, membership determination module 830, pouring coefficient determination module 840, pouring quantity parameter determination module 850, and evaluation module 860 can be combined into one module, or any one of these modules can be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules can be combined with at least part of the functionality of other modules and implemented in one module. According to embodiments of the present invention, at least one of the segmentation module 810, area determination module 820, membership determination module 830, pouring coefficient determination module 840, pouring quantity determination module 850, and evaluation module 860 can be at least partially implemented as hardware circuitry, such as field-programmable gate arrays (FPGAs), programmable logic arrays (PLAs), systems-on-a-chip, systems-on-a-substrate, systems-on-package, application-specific integrated circuits (ASICs), or any other reasonable means of integrating or packaging circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, at least one of the segmentation module 810, area determination module 820, membership degree determination module 830, pouring coefficient determination module 840, pouring volume determination module 850, and evaluation module 860 can be at least partially implemented as a computer program module, which can perform the corresponding function when the computer program module is run.

[0172] Figure 9 This is a structural block diagram of an electronic device suitable for determining the pouring parameters of precast beam concrete defects according to an embodiment of the present invention.

[0173] like Figure 9 As shown, an electronic device 900 according to an embodiment of the present invention includes a processor 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage portion 908 into a random access memory (RAM) 903. The processor 901 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 901 may also include onboard memory for caching purposes. The processor 901 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0174] RAM 903 stores various programs and data required for the operation of electronic device 900. Processor 901, ROM 902, and RAM 903 are interconnected via bus 904. Processor 901 executes various operations of the method flow according to embodiments of the present invention by executing programs in ROM 902 and / or RAM 903. It should be noted that programs may also be stored in one or more memories other than ROM 902 and RAM 903. Processor 901 may also execute various operations of the method flow according to embodiments of the present invention by executing programs stored in one or more memories.

[0175] According to an embodiment of the present invention, the electronic device 900 may further include an input / output (I / O) interface 905, which is also connected to a bus 904. The electronic device 900 may also include one or more of the following components connected to the I / O interface 905: an input section 906 including a keyboard, mouse, etc.; an output section 907 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the I / O interface 905 as needed. A removable medium 911, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 910 as needed so that computer programs read from it can be installed into the storage section 908 as needed.

[0176] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of the present invention.

[0177] According to embodiments of the present invention, a computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of the present invention, a computer-readable storage medium may include ROM 902 and / or RAM 903 and / or one or more memories other than ROM 902 and RAM 903 described above.

[0178] Embodiments of the present invention also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to enable the computer system to implement the method for determining precast beam concrete defect casting parameters provided in the embodiments of the present invention.

[0179] When the computer program is executed by the processor 901, it performs the functions defined in the system / apparatus of this invention. According to embodiments of the invention, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0180] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and downloaded and installed via the communication section 909, and / or installed from a removable medium 911. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0181] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 909, and / or installed from the removable medium 911. When the computer program is executed by the processor 901, it performs the functions defined in the system of this embodiment of the invention. According to embodiments of the invention, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0182] According to embodiments of the present invention, program code for executing the computer programs provided in the embodiments of the present invention can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0183] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0184] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for determining pouring parameters for concrete defects in precast beams, characterized in that, include: The image of the precast beam is acquired, and the pixel region of the concrete surface defect and the confidence level corresponding to each defect category are obtained through the instance segmentation model. The physical area of ​​the concrete surface defect is determined based on the mapping relationship between the pixel region and the actual size of the precast beam. The confidence level of each defect category is used as the defect category membership degree, and the membership degree of the pouring effect is determined based on the pouring effect evaluation value of the previous pouring cycle. Based on the membership degree of the defect category and the membership degree of the pouring effect, the pouring coefficient within this pouring cycle is determined through fuzzy reasoning, including: Based on the membership degree of the defect category and the membership degree of the pouring effect, a preset fuzzy rule is matched to determine the triggered pouring strategy and its corresponding activation intensity. The membership functions corresponding to each of the pouring strategies are pruned according to the activation intensity to obtain multiple pruned output fuzzy sets; The output fuzzy sets are merged to form a comprehensive fuzzy set; The comprehensive fuzzy set is defuzzified using the shape centroid method to determine the adjustment amount of the casting coefficient; The pouring coefficient of the previous pouring cycle is adjusted according to the pouring coefficient adjustment amount to obtain the pouring coefficient of the current pouring cycle. The weight parameters of the concrete used to fill the surface defects of the concrete are determined based on the physical area and the pouring coefficient. Obtain defect images before and after this pouring to determine the evaluation value of the pouring effect for this pouring cycle.

2. The method for determining the pouring parameters for precast beam concrete defects according to claim 1, characterized in that, The process of acquiring defect images before and after the current pouring to determine the evaluation value of the pouring effect for this pouring cycle includes: Obtain a first image of the concrete surface defects before this pouring, and a second image of the defects after this pouring; The area of ​​the first defect before this pouring is determined based on the first defect image, and the first confidence level corresponding to each defect type is determined. The area of ​​the second defect after this pouring is determined based on the second defect image, as well as the second confidence level corresponding to each defect type; The evaluation value of the pouring effect for this pouring cycle is calculated based on the first defect area, the first confidence level, the second defect area, the second confidence level, and the preset defect severity weight vector.

3. The method for determining the pouring parameters for precast beam concrete defects according to claim 1, characterized in that, Also includes: Based on the mapping relationship between the pixel size of the pixel region and the actual size of the precast beam, the pixel coordinates in the pixel region are converted into the local actual coordinates of the concrete surface defects; A global coordinate system for the precast beam is established with the preset point of the precast beam as the origin. The local actual coordinates are translated into the global coordinate system to determine the physical location of the concrete surface defects on the precast beam.

4. The method for determining the pouring parameters for precast beam concrete defects according to claim 3, characterized in that, The determination of the concrete weight parameters for pouring the concrete surface defects based on the physical area and pouring coefficient includes: The concrete surface defects are divided into multiple areas to be poured. Determine the center location and area of ​​each of the areas to be poured; The pouring location is determined by the center position of each of the areas to be poured, and the concrete weight parameters of each of the areas to be poured are determined based on the area of ​​each of the areas to be poured.

5. The method for determining the pouring parameters for precast beam concrete defects according to claim 4, characterized in that, The process of dividing the concrete surface defects into multiple areas to be poured includes: The bounding rectangle of the concrete surface defect is determined based on the pixel region, and the bounding rectangle is divided into equal intervals in the horizontal and vertical directions to form multiple areas to be poured.

6. The method for determining the pouring parameters for precast beam concrete defects according to claim 1, characterized in that, It also includes updating the instance segmentation model based on newly acquired defect images to improve defect recognition accuracy.

7. A device for determining pouring parameters for concrete defects in precast beams, characterized in that, include: The segmentation module is used to acquire images of precast beams and obtain pixel regions of concrete surface defects and confidence levels corresponding to each defect category through an instance segmentation model. An area determination module is used to determine the physical area of ​​the concrete surface defects based on the mapping relationship between the pixel region and the actual size of the precast beam. The membership determination module is used to take the confidence level of each defect category as the defect category membership degree, and to determine the pouring effect membership degree based on the pouring effect evaluation value of the previous pouring cycle. The pouring coefficient determination module is used to determine the pouring coefficient within the current pouring cycle through fuzzy reasoning based on the defect category membership degree and the pouring effect membership degree. The pouring coefficient determination module includes: The activation intensity determination submodule is used to match preset fuzzy rules based on the defect category membership degree and the pouring effect membership degree to determine the triggered pouring strategy and its corresponding activation intensity. The fuzzy set output submodule is used to trim the membership function corresponding to each of the casting strategies according to the activation intensity, so as to obtain multiple trimmed output fuzzy sets. The merging submodule is used to merge the various output fuzzy sets to form a comprehensive fuzzy set; The casting coefficient adjustment amount determination submodule is used to defuzzify the comprehensive fuzzy set according to the shape centroid method in order to determine the casting coefficient adjustment amount; The pouring coefficient determination submodule is used to adjust the pouring coefficient of the previous pouring cycle according to the pouring coefficient adjustment amount to obtain the pouring coefficient of the current pouring cycle. The pouring volume parameter determination module is used to determine the concrete weight parameters for pouring the concrete surface defects based on the physical area and the pouring coefficient. The evaluation module is used to obtain defect images before and after the current pouring to determine the evaluation value of the pouring effect for this pouring cycle.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for determining the pouring parameters of precast beam concrete defects as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for determining the pouring parameters of precast beam concrete defects as described in any one of claims 1 to 6.

Images