An elliptical glass curtain wall defect detection method and system

Abstract

The application discloses an elliptical glass curtain wall defect detection method and system, comprising: constructing a thin model of an elliptical glass curtain wall to be detected; the elliptical glass curtain wall to be detected comprises a plurality of stress points; based on the thin model, stress values of the plurality of stress points are determined; defect perception data of a plurality of target stress points are acquired; the target stress points are stress points with stress values greater than a preset threshold among the plurality of stress points; the defect perception data is subjected to contrast enhancement processing to obtain defect perception data after enhancement; based on the least square method, the defect perception data after enhancement is fitted to obtain a defect data image of the elliptical glass curtain wall to be detected; and the defect data image is subjected to fusion processing to obtain fine defect information of the elliptical glass curtain wall to be detected. The application alleviates the technical problem of high detection error in the prior art.

Description

Technical Field

[0001] This invention relates to the field of glass curtain wall defect detection technology, specifically to a method and system for detecting defects in elliptical glass curtain walls. Background Technology

[0002] With increasing demands for aesthetically pleasing building structures, elliptical glass curtain walls are being used more and more widely. Compared to planar glass curtain walls, elliptical curtain walls have a more uneven stress structure and more concentrated stress points. To ensure the stability of elliptical glass curtain walls in practical applications, researchers have proposed combining multi-sensor fusion detection technology to analyze the real-time state of each stress point on the elliptical glass curtain wall. Using this as a reference, the sensor results from multiple stress points are fused to determine the real-time stability of the elliptical glass curtain wall. Many researchers have proposed multi-sensor fusion detection technology in this regard.

[0003] Reference [1] "Multi-sensor detection pipeline defect data fusion method" Liang Haibo, Cheng Gang, Zhang Zhidong et al. used laser radar to scan the device under test and used the scanning results as the external parameters for nonlinear optimization. The installation position of the multi-sensor was set in combination with the parameter results, and the real-time status of the device under test was continuously identified. Three-dimensional image data was constructed in the form of frame stacking, continuous video frames were processed, and the data was fused to obtain the multi-sensor fusion detection results. Experimental results show that the detection efficiency of this method is low. Reference [2] "Laser and Optoelectronics Progress, 2023, 60(04)" used an intelligent sensing system to fuse multiple sensors and established three detection layers: data layer, representation layer and decision layer. Four sensors in the multi-sensor were selected, and the data information they sensed was directly fused to form the wear signal of the device under test. Quantitative analysis was performed using the interpolation method to obtain the detection results. Experimental results show that the detection error of this detection method is high. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for detecting defects in elliptical glass curtain walls in order to solve at least one of the above-mentioned technical problems.

[0005] In a first aspect, embodiments of the present invention provide a method for detecting defects in an elliptical glass curtain wall, comprising: constructing a thin model of the elliptical glass curtain wall to be detected; the elliptical glass curtain wall to be detected includes multiple stress points; determining the stress values ​​of the multiple stress points based on the thin model; acquiring defect sensing data of multiple target stress points; the target stress points being stress points among the multiple stress points whose stress values ​​are greater than a preset threshold; performing contrast enhancement processing on the defect sensing data to obtain enhanced defect sensing data; fitting the enhanced defect sensing data based on the least squares method to obtain a defect data image of the elliptical glass curtain wall to be detected; and performing fusion processing on the defect data image to obtain detailed defect information of the elliptical glass curtain wall to be detected.

[0006] Furthermore, acquiring defect perception data of multiple target stress points includes: setting sensors on the multiple target stress points; and acquiring defect perception data of the multiple target stress points based on the sensors.

[0007] Further, the defect perception data is subjected to contrast enhancement processing to obtain enhanced defect perception data, including: dividing the multiple target stress points into multiple defect regions according to their positions on the elliptical glass curtain wall to be detected; enhancing the defect perception data of the target stress points in each defect region, and converting the processed defect perception data into grayscale images; wherein each defect region corresponds to a grayscale image; obtaining the edge jitter of the grayscale image corresponding to each defect region; and performing contrast processing on each grayscale image based on the edge jitter to obtain the enhanced defect perception data.

[0008] Further, the enhanced defect perception data is fitted using the least squares method to obtain a defect data image of the elliptical glass curtain wall to be detected, including: fitting the enhanced defect perception data using the least squares method to obtain an initial fitting curve; performing analog-to-digital conversion on the initial fitting curve to obtain a linear relationship between the enhanced defect perception data and the digital quantity; fitting a defect three-dimensional position curve based on defect perception data from different sensors and the linear relationship; and reconstructing the defect data image of the elliptical glass curtain wall to be detected based on the defect three-dimensional position curve.

[0009] Further, the defect data images are fused to obtain detailed defect information of the elliptical glass curtain wall to be detected, including: performing difference judgment on defect data images corresponding to different sensors based on a preset sliding window to obtain a judgment result; performing weight compensation on defect data images corresponding to different sensors to obtain compensated defect data images; and performing fusion processing on the compensated defect data images based on the judgment result to obtain detailed defect information of the elliptical glass curtain wall to be detected.

[0010] Secondly, embodiments of the present invention also provide a defect detection system for elliptical glass curtain walls, comprising: a construction module, a determination module, an acquisition module, a processing module, a fitting module, and a fusion module; wherein, the construction module is used to construct a thin model of the elliptical glass curtain wall to be detected; the elliptical glass curtain wall to be detected includes multiple stress points; the determination module is used to determine the stress values ​​of the multiple stress points based on the thin model; the acquisition module is used to acquire defect perception data of multiple target stress points; the target stress points are the stress points among the multiple stress points whose stress values ​​are greater than a preset threshold; the processing module is used to perform contrast enhancement processing on the defect perception data to obtain enhanced defect perception data; the fitting module is used to fit the enhanced defect perception data based on the least squares method to obtain a defect data image of the elliptical glass curtain wall to be detected; the fusion module is used to perform fusion processing on the defect data image to obtain fine defect information of the elliptical glass curtain wall to be detected.

[0011] Furthermore, the processing module is also configured to: divide the plurality of target stress points into a plurality of defect regions according to their positions on the elliptical glass curtain wall to be detected; enhance the defect perception data of the target stress points in each defect region, and convert the processed defect perception data into a grayscale image; wherein each defect region corresponds to a grayscale image; obtain the edge jitter of the grayscale image corresponding to each defect region; and perform contrast processing on each grayscale image based on the edge jitter to obtain the enhanced defect perception data.

[0012] Furthermore, the fitting module is also used to: fit the enhanced defect perception data based on the least squares method to obtain an initial fitting curve; perform analog-to-digital conversion on the initial fitting curve to obtain a linear relationship between the enhanced defect perception data and the digital quantity; fit a defect three-dimensional position curve based on defect perception data from different sensors and the linear relationship; and reconstruct a defect data image of the elliptical glass curtain wall to be detected based on the defect three-dimensional position curve.

[0013] Furthermore, the fusion module is also used to: perform difference judgment on defect data images corresponding to different sensors based on a preset sliding window to obtain a judgment result; perform weight compensation on defect data images corresponding to different sensors to obtain a compensated defect data image; and perform fusion processing on the compensated defect data image based on the judgment result to obtain fine defect information of the elliptical glass curtain wall to be detected.

[0014] Thirdly, embodiments of the present invention also provide 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 computer program to implement the method described in the first aspect above.

[0015] This invention provides a method and system for detecting defects in elliptical glass curtain walls, comprising: constructing a thin model of the elliptical glass curtain wall to be detected; the elliptical glass curtain wall to be detected includes multiple stress points; determining the stress values ​​of the multiple stress points based on the thin model; acquiring defect sensing data of multiple target stress points; the target stress points are the stress points among the multiple stress points whose stress values ​​are greater than a preset threshold; performing contrast enhancement processing on the defect sensing data to obtain enhanced defect sensing data; fitting the enhanced defect sensing data based on the least squares method to obtain a defect data image of the elliptical glass curtain wall to be detected; and performing fusion processing on the defect data image to obtain detailed defect information of the elliptical glass curtain wall to be detected. This invention alleviates the technical problem of high detection errors existing in the prior art. Attached Figure Description

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

[0017] Figure 1 A flowchart of a defect detection method for an elliptical glass curtain wall provided in an embodiment of the present invention;

[0018] Figure 2 A schematic diagram of a thin model of an elliptical glass curtain wall provided in an embodiment of the present invention;

[0019] Figure 3 This is a schematic diagram illustrating the division of defect areas in an elliptical glass curtain wall to be inspected, provided as an embodiment of the present invention.

[0020] Figure 4 This is a schematic diagram of a multi-sensor defect detection image fusion process provided in an embodiment of the present invention;

[0021] Figure 5 A schematic diagram of sensor installation for an elliptical glass curtain wall to be detected, provided as an embodiment of the present invention;

[0022] Figure 6 This is a schematic diagram of the detection results of a defect in an elliptical glass curtain wall provided in an embodiment of the present invention;

[0023] Figure 7 This is a schematic diagram of the fitting result for the detection of defects in an elliptical glass curtain wall provided in an embodiment of the present invention;

[0024] Figure 8 This is a schematic diagram illustrating the cumulative error results of different methods provided in an embodiment of the present invention;

[0025] Figure 9 This is a schematic diagram of an elliptical glass curtain wall defect detection system provided in an embodiment of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0027] Example 1

[0028] Figure 1 This is a flowchart of a defect detection method for an elliptical glass curtain wall according to an embodiment of the present invention. Figure 1 As shown, the method specifically includes the following steps:

[0029] Step S102: Construct a thin model of the elliptical glass curtain wall to be tested; the elliptical glass curtain wall to be tested includes multiple stress points.

[0030] Step S104: Based on the thin model, determine the stress values ​​at multiple stress points.

[0031] Step S106: Obtain defect perception data of multiple target stress points; the target stress points are the stress points among the multiple stress points whose stress values ​​are greater than a preset threshold.

[0032] Step S108: Perform contrast enhancement processing on the defect perception data to obtain enhanced defect perception data.

[0033] Step S110 fits the enhanced defect perception data based on the least squares method to obtain the defect data image of the elliptical glass curtain wall to be detected.

[0034] Step S112: The defect data image is fused to obtain detailed defect information of the elliptical glass curtain wall to be inspected.

[0035] This invention provides a method for detecting defects in elliptical glass curtain walls. After analyzing the stress values ​​at the stress points of the elliptical glass curtain wall and the detection data signals from the contrast-enhanced grouped sensors to obtain the initial fitting curve of the defect location, the least squares method is used to fit the three-dimensional position curve of the defect location, and the defect detection image is fused to obtain the detection result. The defect detection result obtained by this method has low error and good practical application effect.

[0036] Generally, the thickness of a glass curtain wall is much smaller than its surface area. Therefore, when performing stress point analysis, the effect of the thickness of the elliptical glass panel is ignored. The elliptical glass curtain wall is treated as a thin model, such as... Figure 2 As shown. Among them, Figure 2 This is a schematic diagram of a thin model of an elliptical glass curtain wall provided according to an embodiment of the present invention.

[0037] Specifically, such as Figure 2 As shown, the solid line represents the physical model of the elliptical glass curtain wall, and the dashed line represents the virtual part that needs to be completed for stress point analysis. Here, L represents the central axis of the elliptical glass curtain wall, a represents the major axis, b represents the minor axis, and i represents the selected stress point.

[0038] Draw a tangent line at the stress point i. Extend the perpendicular line of this tangent line to intersect the central axis, forming R2. The angle between this tangent line and the horizontal line is φ. At the intersection point, take 1 / 2φ in the vertical direction and extend it to the surface of the elliptical glass curtain wall to form R1. The distance between this intersection point and the horizontal line of the tangent point is l.

[0039] according to Figure 2 The stress at the stress points of the constructed elliptical glass curtain wall model is calculated. First, the natural angular frequency of the elliptical glass curtain wall is calculated, as shown in formula (1):

[0040]

[0041] In the formula, f represents the natural angular frequency of the elliptical glass curtain wall.

[0042] Combined with f, the stiffness matrix of the force-bearing point i is calculated as shown in formula (2):

[0043]

[0044] The eigenvalues ​​of this matrix are obtained as shown in formula (3):

[0045]

[0046] in, σ represents the eigenvalues ​​of the stiffness matrix, M represents the mass of the elliptical glass curtain wall, and σ represents the eigenvector.

[0047] Based on the influence of stiffness on the stress at stress point i, the stress-stiffness matrix at stress point i is calculated as shown in formula (4):

[0048]

[0049] Where K represents the stress stiffness matrix and μ represents the vector of constraint torque.

[0050] Deconstructing the matrix yields the stress value at point i, as shown in formula (5):

[0051]

[0052] In the formula, D K Let i represent the stress value at the stress point, E represent the elastic modulus, h represent the thickness of the elliptical glass curtain wall, and c represent Poisson's ratio.

[0053] Optionally, in this embodiment of the invention, E is taken as 200 GPa and c is taken as 0.3. Using the above calculation formula, the stress value at the stress point of the elliptical glass curtain wall is obtained.

[0054] Specifically, step S106 includes the following steps:

[0055] Step S1061: Install sensors at multiple target force points;

[0056] Step S1062: Obtain defect perception data of multiple target stress points based on sensors.

[0057] For example, if the preset threshold is 105 MPa, then the analysis of D... K For stress points where the calculated stress value exceeds 105 MPa, sensors are installed at these stress points. After installation, it is checked that all sensors cover the entire elliptical glass curtain wall. Once confirmed to be complete, all sensors are activated, and defect detection data for each point on the elliptical glass curtain wall are acquired simultaneously.

[0058] Specifically, step S108 includes the following steps:

[0059] Step S1081: Divide the multiple target stress points into multiple defect areas according to their positions on the elliptical glass curtain wall to be inspected;

[0060] Step S1082: Enhance the defect perception data of the target stress point in each defect area, and convert the processed defect perception data into a grayscale image; wherein, each defect area corresponds to a grayscale image;

[0061] Step S1083: Obtain the edge jitter of the grayscale image corresponding to each defect region;

[0062] Step S1084: Perform contrast processing on each grayscale image based on edge jitter to obtain enhanced defect perception data.

[0063] Specifically, Figure 3 This is a schematic diagram illustrating the division of defect areas in an elliptical glass curtain wall to be inspected, according to an embodiment of the present invention. Figure 3 As shown, all sensors are divided into four groups according to their positions, with each group measuring 45°, namely groups A, B, C, and D. The defect sensing data obtained is enhanced, and a grayscale image is constructed using the processed data. The median grayscale value of the background region and the median grayscale value of the defect data signal region are then obtained. The difference between the two is the grayscale image contrast value, and the magnitude of this contrast value represents the clarity of the data signal in the defect region. Therefore, to improve the clarity of the detection results, contrast enhancement processing is performed.

[0064] The edge jitter of the defect region signal in the grayscale image is obtained as shown in formula (6):

[0065]

[0066] Where y(z) represents the grayscale contrast value of sensor z, and u represents the edge jitter of the defect region signal in the grayscale image. Optionally, z can take values ​​of 1, 2, 3, and 4.

[0067] Enhanced defect perception data is obtained by contrast processing of grayscale images.

[0068] Specifically, step S110 further includes the following steps:

[0069] Step S1101: Fit the enhanced defect perception data based on the least squares method to obtain the initial fitting curve;

[0070] Step S1102: Perform analog-to-digital conversion on the initial fitting curve to obtain a linear relationship between the enhanced defect perception data and the digital quantity;

[0071] Step S1103: Based on the defect perception data from different sensors and the linear relationship, fit the three-dimensional position curve of the defect;

[0072] Step S1104: Based on the defect three-dimensional position curve, reconstruct the defect data image of the elliptical glass curtain wall to be inspected.

[0073] Specifically, the least squares method is used to fit the enhanced defect perception data to obtain the initial fitting curve of the defect area of ​​the elliptical glass curtain wall to be detected, as shown in formula (7):

[0074] Y = 1.334667D K +1.001300 (7)

[0075] In this embodiment of the invention, the accuracy of the defect data curve fitting is set to level 0.2.

[0076] A / D conversion (i.e., analog-to-digital conversion) is performed on Y. During the conversion process, a linear relationship between the defect data signal and the digital quantity is obtained, as shown in formula (8):

[0077] g=αy(z)+βY (8)

[0078] Where g represents a linear relationship, and α and β represent two coefficients in the A / D conversion process.

[0079] Using formula (7), the values ​​of coefficients α and β when Y is 1 and 5 respectively are calculated, and the results are shown in formulas (9) and (10):

[0080] ―0.000974=α×0+β (9)

[0081] 2.996028=α×27648+β (10)

[0082] The final calculated coefficient α is 0.000208398, and the coefficient β is -0.000974. The images of the elliptical glass curtain wall defects sensed by sensors at different locations are relatively planar. Based on this result, a three-dimensional defect location curve is fitted, as shown in formula (11):

[0083]

[0084] Where A represents the final result of the improved least squares curve fitting, and n represents the amount of defective data in the fitting.

[0085] Based on the calculation results of formula (11), the three-dimensional data image of the defects of the elliptical glass curtain wall is reconstructed.

[0086] Specifically, step S112 further includes the following steps:

[0087] Step S1121: Based on a preset sliding window, perform difference judgment on the defect data images corresponding to different sensors to obtain the judgment result;

[0088] Step S1122: Perform weight compensation on the defect data images corresponding to different sensors to obtain the compensated defect data images;

[0089] Step S1123: Based on the judgment result, the compensated defect data image is fused to obtain detailed defect information of the elliptical glass curtain wall to be detected.

[0090] The defect data images obtained based on least squares curve fitting are fused to fuse images perceived by multiple sensors, resulting in refined defect information. The fusion process of multi-sensor defect detection images is as follows: Figure 4 As shown.

[0091] Specifically, in the defect image of the elliptical glass curtain wall, the defect area exhibits abrupt peaks. Based on this characteristic, when performing defect image fusion, this embodiment of the invention sets a sliding window (i.e., a preset sliding window) in the image. The grayscale value of each pixel inside the sliding window is calculated and compared with the grayscale value of the pixel at the center point of the sliding window. The maximum difference in the comparison results is used for judgment, as shown in formula (12):

[0092]

[0093] Where G represents the decision result, △T represents the set of gray value differences between two points, Q represents the center point of the sliding window, q represents the point inside the sliding window with the maximum gray value difference, and η represents the decision threshold.

[0094] During the decision-making process, the value of the decision factor directly affects the resolution of the fused defect image. Therefore, it is necessary to calculate the value of the decision factor, as shown in formula (13):

[0095]

[0096] Where ρ represents the decision factor, and I(ρ) represents the ratio of grayscale pixels to total pixels when the decision factor is ρ.

[0097] After the judgment is completed, weight compensation is performed on the defect detection images of the elliptical glass curtain wall obtained by different sensors. In this embodiment of the invention, this step is accomplished based on the compensation of image entropy and standard deviation, as shown in formula (14):

[0098]

[0099] Where W represents the weight value that needs to be compensated, t i v represents the image entropy at the point of force application i. i This represents the standard deviation of the defect image at point i under stress.

[0100] After the weight values ​​are compensated, defect detection image fusion can be performed. The final decision function for image fusion is obtained based on formula (12), as shown in formula (15):

[0101] P(q,Q)=G1(q,Q)·G2(q,Q) (15)

[0102] Where P(q,Q) represents the image fusion decision, G1(q,Q) represents the decision result without calculating the decision factor, and G2(q,Q) represents the decision result after calculating the optimal decision factor.

[0103] Based on the above calculations, the image fusion decision is as follows:

[0104] (1) P(q,Q)=1, perform fusion at the maximum pixel value;

[0105] (2) P(q,Q)=0 and G1(q,Q)=1, merge at the center pixel of the sliding window;

[0106] (3) P(q,Q)=0 and G2(q,Q)=1, merge at the point with the maximum gray value difference.

[0107] Different defect detection image fusion decisions are made based on different results to complete the multi-sensor fusion detection of defects in elliptical glass curtain walls.

[0108] Example 2

[0109] To verify the feasibility of the method proposed in this invention, a simulation experiment was designed to verify its effectiveness. This experiment was conducted on an optical platform.

[0110] First, prepare several elliptical glass pieces. Based on the light transmittance of glass, spray a layer of matte white paint onto all elliptical glass pieces to create a diffuse reflection surface. Divide the treated elliptical glass pieces into different experimental groups according to the conventional form of glass curtain walls, ensuring that the shape, number, and size of the glass curtain walls in each experimental group are consistent. Install sensors of the same model at all stress points of each group of glass curtain walls to ensure that the state of each point on the glass curtain wall is included in the sensing range, such as... Figure 5 As shown.

[0111] The sensor parameters used in this experiment are shown in Table 1.

[0112] Table 1 Sensor parameter configuration

[0113]

[0114]

[0115] Connect the sensor's data transmission part to a computer running Windows 10 to display the detection results in the form of images.

[0116] Defect detection was conducted on the elliptical glass curtain wall experimental group as shown above, and the experimental results are as follows: Figure 6 As shown.

[0117] like Figure 6 As shown, three defects were detected in the elliptical glass curtain wall, and each defect exhibited a different state of damage. Computer fitting determined the locations of the three defects within the glass curtain wall as follows: Figure 7 As shown.

[0118] according to Figure 7 It can be preliminarily determined that the detection results of the proposed method are relatively reasonable, indicating the feasibility of the proposed method.

[0119] To more intuitively demonstrate the effectiveness of the proposed method, the cumulative error of multiple detection results is used as the analytical index of this experiment. As shown in formula (16):

[0120] δ = 10 2 Ssin(δ θ (16)

[0121] Where δ represents the cumulative error of multiple defect detection results, S represents the total area of ​​the elliptical glass curtain wall, and δ θ This indicates the elevation angle error of the sensor.

[0122] The calculated cumulative error directly reflects the performance of multi-sensor fusion detection of defects in elliptical glass curtain walls. The smaller the calculated cumulative error, the better the detection results of the method.

[0123] To demonstrate the effectiveness of the proposed method, the multi-sensor fusion detection methods proposed in references [1] and [2] (mentioned in the background section) were applied to conduct defect detection experiments on the elliptical glass curtain wall constructed in this experiment. A total of 10 groups of curtain walls were detected, and the cumulative error of the 10 defect detection results was calculated. This experiment was conducted 5 times, and the cumulative error of each detection result was recorded during the experiment. The multi-sensor fusion detection results of defects in the elliptical glass curtain wall under different methods are as follows: Figure 8 As shown.

[0124] Depend on Figure 8It can be seen that the cumulative error of the defect detection method for elliptical glass curtain walls provided in this embodiment of the invention is smaller than that of the other two methods. In five rounds of testing experiments, the cumulative error of the multi-sensor fusion detection result of the defect detection method for elliptical glass curtain walls provided in this embodiment of the invention never exceeded 0.1, with an average cumulative error of only 0.07. In contrast, the average cumulative errors of the other two methods reached 0.54 and 0.68, respectively. The comparative experimental results show that the defect detection method for elliptical glass curtain walls provided in this embodiment of the invention obtains a smaller error in the multi-sensor fusion detection result for elliptical glass curtain walls, exhibits superior detection performance, and has high practical application value.

[0125] In the development of the construction industry, the application of elliptical glass curtain walls is becoming increasingly widespread. To ensure the safety of elliptical glass curtain walls, defect detection is necessary. This invention provides a method for defect detection in elliptical glass curtain walls. Experimental results show that the defect detection results obtained using the proposed method have low error and good practical application effects.

[0126] Example 3

[0127] Figure 9 This is a schematic diagram of an elliptical glass curtain wall defect detection system according to an embodiment of the present invention. Figure 9 As shown, the system includes: a construction module 10, a determination module 20, an acquisition module 30, a processing module 40, a fitting module 50, and a fusion module 60.

[0128] Specifically, module 10 is used to construct a thin model of the elliptical glass curtain wall to be tested; the elliptical glass curtain wall to be tested includes multiple stress points.

[0129] Module 20 is used to determine the stress values ​​at multiple stress points based on a thin model.

[0130] The acquisition module 30 is used to acquire defect perception data of multiple target stress points; the target stress points are the stress points among the multiple stress points whose stress values ​​are greater than a preset threshold.

[0131] Specifically, sensors are set at multiple target stress points; defect perception data of multiple target stress points are obtained based on the sensors.

[0132] The processing module 40 is used to perform contrast enhancement processing on the defect perception data to obtain enhanced defect perception data.

[0133] The fitting module 50 is used to fit the enhanced defect perception data based on the least squares method to obtain the defect data image of the elliptical glass curtain wall to be detected.

[0134] The fusion module 60 is used to fuse the defect data images to obtain detailed defect information of the elliptical glass curtain wall to be inspected.

[0135] This invention provides a defect detection system for elliptical glass curtain walls. After analyzing the stress values ​​at the stress points of the elliptical glass curtain wall and the detection data signals from the contrast-enhanced grouped sensors to obtain the initial fitting curve of the defect location, the least squares method is used to fit the three-dimensional position curve of the defect location, and the defect detection image is fused to obtain the detection result. The defect detection result obtained by this method has low error and good practical application effect.

[0136] Specifically, the processing module 40 is also used for:

[0137] The multiple target stress points are divided into multiple defect areas according to their positions on the elliptical glass curtain wall to be inspected;

[0138] The defect perception data of the target stress point within each defect region is enhanced, and the processed defect perception data is converted into a grayscale image; where each defect region corresponds to a grayscale image.

[0139] Obtain the edge jitter of the grayscale image corresponding to each defect region;

[0140] Each grayscale image is contrast-processed based on edge jitter to obtain enhanced defect perception data.

[0141] Specifically, the fitting module 50 is also used for:

[0142] The enhanced defect perception data is fitted using the least squares method to obtain an initial fitting curve;

[0143] The initial fitted curve is converted from analog to digital to obtain a linear relationship between the enhanced defect perception data and the digital quantity.

[0144] Based on defect perception data from different sensors and linear relationships, a three-dimensional defect location curve is fitted.

[0145] Based on the 3D location curve of the defect, the defect data image of the elliptical glass curtain wall to be inspected is reconstructed.

[0146] Specifically, the fusion module 60 is also used for:

[0147] Based on a preset sliding window, the defect data images corresponding to different sensors are used to make a difference judgment to obtain the judgment result;

[0148] Weight compensation is performed on the defect data images corresponding to different sensors to obtain the compensated defect data images;

[0149] Based on the judgment, the compensated defect data images are fused to obtain detailed defect information of the elliptical glass curtain wall to be inspected.

[0150] This 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 computer program to implement the method described in Embodiment 1 above.

[0151] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0152] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for detecting defects in elliptical glass curtain walls, characterized in that, include: A thin model of the elliptical glass curtain wall to be tested is constructed; the elliptical glass curtain wall to be tested includes multiple stress points; Based on the thin model, the stress values ​​of the plurality of stress points are determined; Acquire defect perception data from multiple target stress points; The target stress point is the stress point among the plurality of stress points whose stress value is greater than a preset threshold; The defect perception data is subjected to contrast enhancement processing to obtain enhanced defect perception data; The enhanced defect perception data is fitted using the least squares method to obtain the defect data image of the elliptical glass curtain wall to be detected. The defect data images are fused to obtain detailed defect information of the elliptical glass curtain wall to be detected; The enhanced defect perception data is fitted using the least squares method to obtain a defect data image of the elliptical glass curtain wall to be detected, including: The enhanced defect perception data is fitted using the least squares method to obtain an initial fitting curve; The initial fitting curve is converted from analog to digital to obtain a linear relationship between the enhanced defect perception data and the digital quantity. Based on defect perception data from different sensors and the linear relationship described above, a three-dimensional defect position curve is fitted. Based on the defect three-dimensional position curve, the defect data image of the elliptical glass curtain wall to be detected is reconstructed; The defect data images are fused to obtain detailed defect information of the elliptical glass curtain wall to be inspected, including: Based on a preset sliding window, the defect data images corresponding to different sensors are used to make a difference judgment to obtain the judgment result; Weight compensation is performed on the defect data images corresponding to different sensors to obtain the compensated defect data images; Based on the judgment result, the compensated defect data image is fused to obtain detailed defect information of the elliptical glass curtain wall to be detected.

2. The method according to claim 1, characterized in that: Acquire defect perception data from multiple target stress points, including: Sensors are installed at the multiple target force points; Defect perception data of the multiple target stress points are acquired based on the sensor.

3. The method according to claim 1, characterized in that: The defect-sensing data is subjected to contrast enhancement processing to obtain enhanced defect-sensing data, including: The multiple target stress points are divided into multiple defect areas according to their positions on the elliptical glass curtain wall to be inspected; The defect perception data of the target stress point within each defect region is enhanced, and the processed defect perception data is converted into a grayscale image; where each defect region corresponds to a grayscale image. Obtain the edge jitter of the grayscale image corresponding to each defect region; Based on the edge jitter, contrast processing is performed on each grayscale image to obtain the enhanced defect perception data.

4. A defect detection system for elliptical glass curtain walls, characterized in that, include: The module includes a construction module, a determination module, an acquisition module, a processing module, a fitting module, and a fusion module; among which, The construction module is used to construct a thin model of the elliptical glass curtain wall to be tested; the elliptical glass curtain wall to be tested includes multiple stress points; The determining module is used to determine the stress values ​​of the plurality of stress points based on the thin model; The acquisition module is used to acquire defect perception data of multiple target stress points; the target stress points are the stress points among the multiple stress points whose stress values ​​are greater than a preset threshold. The processing module is used to perform contrast enhancement processing on the defect perception data to obtain enhanced defect perception data. The fitting module is used to fit the enhanced defect perception data based on the least squares method to obtain the defect data image of the elliptical glass curtain wall to be detected. The fusion module is used to perform fusion processing on the defect data image to obtain detailed defect information of the elliptical glass curtain wall to be detected. The fitting module is also used for: The enhanced defect perception data is fitted using the least squares method to obtain an initial fitting curve; The initial fitting curve is converted from analog to digital to obtain a linear relationship between the enhanced defect perception data and the digital quantity. Based on defect perception data from different sensors and the linear relationship described above, a three-dimensional defect position curve is fitted. Based on the defect three-dimensional position curve, the defect data image of the elliptical glass curtain wall to be detected is reconstructed; The fusion module is also used for: Based on a preset sliding window, the defect data images corresponding to different sensors are used to make a difference judgment to obtain the judgment result; Weight compensation is performed on the defect data images corresponding to different sensors to obtain the compensated defect data images; Based on the judgment result, the compensated defect data image is fused to obtain detailed defect information of the elliptical glass curtain wall to be detected.

5. The system according to claim 4, characterized in that: The processing module is further configured to: The multiple target stress points are divided into multiple defect areas according to their positions on the elliptical glass curtain wall to be inspected; The defect perception data of the target stress point within each defect region is enhanced, and the processed defect perception data is converted into a grayscale image; where each defect region corresponds to a grayscale image. Obtain the edge jitter of the grayscale image corresponding to each defect region; Based on the edge jitter, contrast processing is performed on each grayscale image to obtain the enhanced defect perception data.

6. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method as claimed in any one of claims 1-3.

Images