A building structure deformation measurement method, system and device

Abstract

The application provides a kind of building structure deformation measurement method, system and device, related to building structure deformation measurement technical field.Measurement method includes: collection laser point physical coordinates, the angle of laser emission direction and world coordinate system and building image;After the pretreatment of building image, extract laser point pixel coordinates;After calculating the relative coordinates of material surface laser point, solve the homographic transformation matrix, correct the perspective of building image, obtain the orthographic projection image and obtain the conversion coefficient between pixel and actual distance;The pretreated building image is input into the pre-trained semantic segmentation model, and the beam body segmentation mask and the beam body edge curve are obtained;The actual span of the beam body is obtained by conversion coefficient conversion.This scheme realizes three-dimensional movement through electric drive platform, and can adjust the positioning position of laser points according to the size and position of building structure, adapt to buildings of different heights and angles, and solve the limitation that the prior art can only measure specific types or sizes of building structures.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of building structure deformation measurement, in particular to a building structure deformation measurement method, system and device. BACKGROUND

[0002] Building structures will produce various deformations during long-term use, among which deflection is an important indicator for measuring the safety of building structures and plays a key role in evaluating structural safety. Traditional deflection measurement methods mainly include two categories: contact measurement and non-contact measurement. The contact measurement method needs to be in direct contact with the measured structure, which is not only complex to implement but also has limited measurement points. Although the non-contact measurement method avoids direct contact, it has limited measurement range and cannot obtain multi-dimensional information simultaneously.

[0003] With the development of computer vision and laser ranging technology, non-contact deflection measurement based on optical images has gradually become a research hotspot. This method captures structure images through a camera and calculates the displacement and deformation of the structure by combining image processing algorithms. It has the advantages of non-contact, multi-point measurement, and large amount of information, but still has many shortcomings and cannot meet the measurement needs of building structure deformation.

[0004] The journal paper entitled "Research on Image Processing Algorithm for Laser Reference Bridge Deflection Detection" (Zhao Hongwei et al., Instrument Technology and Sensors, 2019) discloses using a pyramid model to obtain a region of interest (ROI) and construct a multi-layer target image pyramid. After detecting the light spot area in the top layer image, it is mapped back to the original image. The elliptical fitting method is used to realize sub-pixel level positioning of the laser spot center. Through steps such as traversing boundary points, segmenting sub-regions, randomly selecting boundary points, and solving the ellipse equation by least squares method, high-precision positioning is achieved. This scheme requires the installation of a target on the measured structure, increasing the system complexity, and is sensitive to environmental light interference and atmospheric conditions. Moreover, it is only suitable for bridge deflection detection.

[0005] The paper entitled "Research and Application of Deflection Measurement Method Based on Optical Image" (Feng Jianhao, Jiangsu University, 2021) discloses the establishment of an off-axis digital image correlation method considering off-axis angle, which can realize long-distance non-contact deflection measurement of bridges. The influence of camera internal and external parameters and environmental factors on deflection measurement accuracy is systematically analyzed. An image rectification method based on SIFT algorithm is established, which effectively eliminates the measurement error caused by the change of the unmanned aerial vehicle's own attitude when measuring bridge deflection. The measurement accuracy of this scheme will be greatly reduced as the distance increases, and the light intensity requirement is also relatively high. When the light intensity is weak, the measurement error is large, and the surface of the measured structure needs to be marked with speckle marks. Moreover, the calculation complexity is high.

[0006] Patent application with publication number CN107588913B and title of Bridge deflection detection system and detection method discloses a target design of setting four black solid circles on a white background plate in order to image recognition and positioning; gimbal control supports horizontal rotation in order to accurately record and control the gimbal position; the accurate distance from the telescope optical system to the target is calculated through geometric relationship. This scheme also needs to set targets at multiple measuring points of the bridge, which is not only complex to install, but also needs time for gimbal rotation, affecting real-time performance, and the system is large in size, not convenient to carry and install, and is only suitable for bridge deflection detection.

[0007] Patent application with publication number CN109631787B and title of Transmission type target image spot center detection method and bridge deflection image type detection device discloses that all boundary points are divided into multiple sub-regions according to coordinate positions, boundary points are randomly selected from the sub-regions and it is ensured that there is no collinear point; the coefficients in the elliptic equation are solved according to the selected boundary points and by using the least square method; the coordinate calculation is performed by traversing the boundary points to calculate the residual sum of squares, the optimal fitting ellipse is determined and the center coordinates are calculated. This scheme also needs to install a transmission type target on the measured structure, is sensitive to environmental light interference, has high algorithm complexity, and is complex to install and debug, and is only suitable for bridge deflection detection.

[0008] Patent application with publication number CN105067208B and title of A multi-channel bridge deflection monitoring method based on CCD discloses that the displacement is obtained by calculating the difference between the center coordinates of the sub-regions before and after deformation; the sub-pixel search method adopts a quadratic surface fitting method, a local quadratic surface is formed by the element with the largest absolute value in the correlation coefficient matrix and the pixel points around the element, the unknown coefficient is fitted to calculate the sub-pixel position; the deflection is calculated by converting the pixel unit into actual unit through the scale calibration method. This scheme needs to set targets at multiple measuring points of the bridge, which is not only complex in system wiring and high in installation and maintenance cost, but also has high requirements for light conditions, needs additional lighting equipment, and is greatly affected by environmental factors, and is only suitable for bridge deflection detection. SUMMARY

[0009] The present application aims to provide a building structure deformation measurement method, system and device to solve at least one of the above technical problems in the prior art.

[0010] In a first aspect, to solve the above technical problems, the present application provides a building structure deformation measurement method, comprising the following steps:

[0011] Step 1, set up the size measurement device and perform preliminary calibration, specifically including laser range finder, monocular camera and electric drive platform; the monocular camera is located at the center of the table top of the electric drive platform, and the four laser range finders are located at the four corners of the table top of the electric drive platform; the initial direction of laser emission is perpendicular to the table top of the electric drive platform and parallel to the optical axis of the monocular camera; the laser range finder includes angle (e.g. pitch and rotation) motor and angle sensor for controlling the laser emission angle.

[0012] In a feasible implementation, the preliminary calibration specifically includes:

[0013] The monocular camera is calibrated by Zhang Zhengyou plane calibration method to obtain and store internal and external parameters;

[0014] The table top of the electric drive platform is moved to the standard calibration area, and the initial physical coordinates of the four laser range finders are calibrated by the standard size calibration board (building with known length and width), and the installation offset of each laser range finder relative to the center of the table top is recorded;

[0015] The table top of the electric drive platform is moved by a preset distance along the X, Y and Z directions respectively, and the laser range finder readings are compared with the actual moving distance of the table top to establish a three-degree-of-freedom position correction model, including the current positioning coordinates of the table top, for subsequent coordinate correction;

[0016] The angle motor and angle sensor are calibrated to establish an angle measurement model, including the angle between the laser emission direction and the world coordinate system.

[0017] In this way, the foundation for subsequent accurate acquisition of image and laser range finder position is laid, and at the same time, the laser range finder is kept as parallel to the ground as possible to reduce the subsequent measurement error.

[0018] In a feasible implementation, the electric drive platform includes servo motor and / or stepper motor and / or linear module to realize three-dimensional movement and accurate positioning of the table top.

[0019] Step 2, move the table top of the electric drive platform, rotate the angle motor, and make the laser points (light spots) of each laser range finder project to the corresponding corner points of the building facade to be measured, then collect the laser ranging data after preliminary calibration as the laser point physical coordinates, collect the angle between the laser emission direction and the world coordinate system through the angle sensor, and collect the building image through the monocular camera; the building image includes the laser point and the building facade.

[0020] In a feasible implementation, the laser point physical coordinates belong to the world coordinate system and can be respectively denoted as (x1, y1, z1), (x2, y2, z2), (x3, y3, z3) and (x4, y4, z4). ​​​​

[0021] Step 3, after pre-processing the building image, the laser point pixel coordinates are extracted to provide high-precision (±1 pixel) image reference for subsequent calculations.

[0022] In a feasible implementation, the step 3 specifically includes:

[0023] Step 31, through noise suppression, HSV color domain conversion, mask extraction and mask morphology optimization, a laser point candidate region mask is obtained.

[0024] In a feasible implementation, the step 31 specifically includes:

[0025] Step 311, the original image of the building image is smoothed by a Gaussian convolution kernel (Gaussian function standard deviation 1.2) of a preset size, to obtain a smoothed image, so that the adjacent pixels are weighted and averaged through the normal distribution characteristics, thereby effectively suppressing Gaussian noise and salt and pepper noise;

[0026] Step 312, the original image is selected as a guide image, and the smoothed image is guided filtered (filter radius is 5, and regularization parameter is 100) to obtain a Gaussian filter image, so that the high-frequency interference caused by the building surface texture is eliminated while the gray difference between the laser point and the background is retained, thereby ensuring that the laser point edge is clear and not blurred;

[0027] Step 313, the Gaussian filter image is converted to the HSV color domain, and a V channel threshold range is set to screen out a high-brightness region from the Gaussian filter image;

[0028] Step 314, an S channel threshold range is set to eliminate the white background reflection points in the Gaussian filter image;

[0029] Step 315, a logical AND operation is performed on the screening results of the V channel and the S channel to generate a binary mask containing only the laser point candidate region, thereby excluding more than 95% of non-target pixels and greatly reducing the operation amount of subsequent processing;

[0030] Step 316, morphological closing operation (3*3 rectangular structure element) is performed on the binary mask to fill the (micro) holes inside the laser point (caused by noise); and morphological opening operation (3*3 rectangular structure element) is performed to remove the noise points attached to the edge of the mask, to obtain a laser point candidate region mask (with complete contour and pure inside).

[0031] Step 32, based on the laser point candidate region mask, after extracting the candidate region, the laser point region image is determined through edge detection and contour feature screening, to avoid interference such as workshop light reflection and building surface marker points.

[0032] In an implementation, the step 32 specifically includes:

[0033] Step 321, edge extraction is performed on the Gaussian filtered image processed by the laser point candidate region mask to obtain a candidate region, and then based on a preset double threshold, a continuous edge contour of the laser point is extracted through Gaussian gradient calculation, non-maximum suppression and hysteresis threshold processing, so as to ensure that the edge is not broken and redundant.

[0034] Step 322, after extracting all contours of the candidate region (by the findContours function of the OpenCV tool), based on the geometric features (circular or approximately circular) of the laser point, a screening condition is set to screen out the laser point region image.

[0035] In an implementation, the screening condition includes:

[0036] Screening condition 1, according to the actual size of the laser point (the pixel area after imaging is usually between 50-200 pixels), an area threshold range is set to eliminate small noise contours and large background interference regions.

[0037] Screening condition 2, according to the calculation formula of the circularity , wherein, represents the contour area, represents the contour perimeter, and a circularity threshold range is set to eliminate irregular non-circular contours. Screening condition 3, the ratio of the minimum circumscribed rectangle of the contour to the contour area is calculated, and the contour with a ratio ≤ a preset proportion threshold is screened out, so as to further exclude non-laser point interference.

[0038] Step 33, for the laser point region image, a center pixel coordinate is solved through a fitting algorithm, and the final center pixel coordinate is obtained through multi-frame average optimization, so as to improve the stability of the center pixel coordinate.

[0039] In an implementation, the step 33 specifically includes:

[0040] Step 331, a least square method is used to perform circle fitting on the edge contour of the laser point to solve the center coordinate, and the specific calculation formula includes:

[0041]

[0042] ; wherein,

[0043] represents the center coordinate; represents the radius of the circle;

[0044] ​Using all pixels on the edge contour as samples, an error function is constructed. By minimizing the error function through partial derivatives, the coordinates of the circle center and the circle radius are obtained. The specific expressions include:

[0045] ;

[0046] in, Indicates the first edge contour The coordinates of each pixel;

[0047] Step 332: Since the building image can be multiple frames taken under the same conditions, and the above steps are performed separately to obtain multiple sets of center coordinates, the Grubbs criterion is used to remove abnormal center coordinates (i.e., center coordinates that deviate from the mean by more than 2 standard deviations) from the multiple sets of center coordinates from the multiple frames. The arithmetic mean of the remaining center coordinates is then taken to obtain the four (precise) laser point center pixel coordinates. ( ), ( ), ( )and ( ).

[0048] Step 4: Based on the physical coordinates of the laser point and the center pixel coordinates of the laser point, calculate the relative coordinates of the laser point on the object surface, solve the homography transformation matrix, perform perspective correction on the building image, obtain the orthophoto image, and obtain the conversion coefficient between the pixel and the actual distance. This eliminates the errors caused by the perspective distortion of the monocular camera and the non-ideal plane of the building surface, establishes a high-precision mapping relationship between the pixel and the actual distance, and provides a benchmark for subsequent measurements.

[0049] In one feasible implementation, step 4 specifically includes:

[0050] Step 41: Based on the previous calibration data and laser ranging data, solve for the physical coordinates of the laser point (i.e., the precise position of the laser point in the world coordinate system) through spatial geometric calculations. The specific calculation formula includes:

[0051] ;

[0052] ;

[0053] ;

[0054] in, , , Indicates the first Physical coordinates of each laser point; , , represents the current positioning coordinates of the table top; , represents the installation offset of the first laser range finder relative to the center of the table top; represents the installation offset of the first laser range finder relative to the center of the table top; represents the real-time collected laser ranging data of the first laser range finder; represents the real-time collected laser ranging data of the first laser range finder; , , represents the angle between the laser emission direction and the world coordinate system.

[0055] Step 42, defining the local coordinate system of the object surface ; wherein the projection point of the upper left corner laser point ( ) is taken as the origin , axis is along the building facade horizontally to the right (i.e. parallel to the world coordinate system axis); axis is vertically downward along the building facade (i.e. parallel to the world coordinate system axis); axis is along the normal vector direction of the building facade (which can be fitted by the physical coordinates of the four laser points);

[0056] Project the physical coordinates of the four laser points into the plane of the local coordinate system of the object surface, to obtain the relative coordinates of the object surface laser points, and correct the horizontal projection error caused by the inclination of the axis (small inclination of the building surface), the specific expression including:

[0057] The first projection point , i.e. the origin , has a relative coordinate of ;

[0058] The relative coordinate expression of the second projection point includes:

[0059] ;

[0060] ;

[0061] The relative coordinate expression of the third projection point includes:

[0062] ;

[0063] ;

[0064] The relative coordinate expression of the fourth projection point includes:

[0065] ;

[0066] .

[0067] Step 43, solving homographic transformation matrix, for calculating any point in object plane local coordinate system, corresponding pixel point in image pixel coordinate system, in order to quickly complete mutual conversion between two types of coordinates, providing basis for subsequent image measurement, target positioning, etc., specific expression includes:

[0068] ;

[0069] wherein, , represents object plane laser point relative coordinate; , represents laser point center pixel coordinate; represents 3*3 homographic matrix, specific expression is:

[0070] ;

[0071] wherein, represents matrix element (total 9 elements, but due to homogeneity, actually needs to solve 8 unknown elements);

[0072] expanded as non-homogeneous form specific expression includes:

[0073] ;

[0074] Thus, using object plane laser point relative coordinate and laser point center pixel coordinate between known point pairs, linear equation set can be constructed, that is, to solve (4 groups of known point pairs, 8 linear equations are constructed to solve 8 unknown elements).

[0075] Step 44, based on homographic transformation matrix, through setting correction image size and inverse perspective transformation (in OpenCV vision library), perspective correction is carried out on building image, specific expression includes:

[0076] ;

[0077] wherein, represents inverse matrix of , used for reversing image to object plane transformation to object plane to image orthographic mapping, so as to eliminate perspective distortion; and respectively represent and corrected coordinates.

[0078] Preferably, the correction image size refers to first setting In particular, the maximum relative coordinate is calculated, specifically including:

[0079] Horizontal maximum relative coordinate , i.e. the horizontal distribution range covering all laser points, specifically expressed as: ;

[0080] Vertical maximum relative coordinate , i.e. the vertical distribution range covering all laser points, specifically expressed as: ;

[0081] Then, according to the measurement accuracy requirement (e.g. 0.1 mm / pixel), the maximum relative coordinate is converted into the pixel number, specifically including:

[0082] Horizontal pixel number , specifically expressed as: , wherein, represents the rounding operation;

[0083] Vertical pixel number , specifically expressed as: ;

[0084] In this way, the image size after correction can be matched with the laser point distribution of the local coordinate system of the object surface, ensuring complete coverage of the object surface area and reasonable pixel accuracy.

[0085] Preferably, the inverse perspective transformation can be realized by the warpPerspective function, specifically including: taking the preprocessed building image as the input image; taking as the transformation matrix; taking * as the output size; taking the bicubic interpolation algorithm as the interpolation algorithm; and calling the warpPerspective function to obtain the orthographic projection image (without perspective distortion).

[0086] Step 45, calculating the conversion coefficient between the pixel and the actual distance, specifically including:

[0087] .

[0088] Preferably, the step 45 specifically includes:

[0089] Step 451, calculating the actual distance mean value of the horizontal laser point and the pixel distance mean value , specifically including:

[0090] ;

[0091] ;

[0092] Calculate the average of the actual distance of the vertical laser point and the average of the pixel distance , the specific expression includes:

[0093] ;

[0094] ;

[0095] Step 452, calculate the conversion coefficient between pixels and actual distance , unit: mm / pixel, the specific expression includes:

[0096] ;

[0097] In this way, when the pixel distance in each direction is known, the actual distance can be quickly obtained by directly multiplying , and vice versa.

[0098] Step 5, input the preprocessed building image into the pre-trained semantic segmentation model to obtain the beam body segmentation mask and the beam body edge curve; based on the beam body segmentation mask, the pixel coordinates of the two endpoints a and c of the beam body edge curve are extracted, the straight line segment ac is obtained, and the pixel coordinates of the midpoint b of the straight line segment ac are calculated, so as to obtain the accurate positions of the three feature points, and complete the automatic identification and feature point extraction of the beam body region.

[0099] In a feasible implementation, the specific training method of the semantic segmentation model includes: after preprocessing different sizes and different angles of building images, labeling the beam body region and the beam body edge curve, generating a pixel-level mask label as a training data set; training the semantic segmentation model, the input is the preprocessed building image, and the output is the mask label; through back propagation, the model parameters are optimized to have the ability to accurately segment the beam body region and extract the beam body edge curve.

[0100] Step 6, calculate the length (pixel number) of the straight line segment ac, and convert to obtain the actual span of the beam body through the conversion coefficient; similarly, convert to obtain the height difference between the endpoint a and the midpoint b , the height difference between the endpoint a and the endpoint c , the horizontal distance between the endpoint a and the midpoint b and the horizontal distance between the midpoint b and the endpoint c , and calculate the deflection value , the specific formula is:

[0101] ;

[0102] In this way, the actual span and the deflection value are combined with the building type query to obtain the corresponding building specification, so that compliance can be accurately judged.

[0103] In an embodiment, the conversion in step 6 refers to multiplying the pixel distance of the measured straight line segment by the conversion coefficient.

[0104] In a second aspect, based on the same inventive concept, the present application also provides a building structure deformation measurement system, comprising a data receiving module, a data processing module and a result generating module.

[0105] The data receiving module is configured to receive laser ranging data as physical coordinates of laser points, an angle between a laser emission direction and a world coordinate system, and a building image.

[0106] The data processing module comprises a preprocessing unit, a perspective correction unit, a semantic segmentation unit and a size measurement unit.

[0107] The preprocessing unit is configured to preprocess the building image and extract pixel coordinates of laser points.

[0108] The perspective correction unit is configured to calculate relative coordinates of the laser points on the object surface based on the physical coordinates of the laser points and the pixel coordinates of the laser point centers, solve a homographic transformation matrix, correct the perspective of the building image, obtain an orthographic projection image and obtain a conversion coefficient between the pixel and the actual distance.

[0109] The semantic segmentation unit is configured to input the preprocessed building image into a pre-trained semantic segmentation model to obtain a beam segmentation mask and a beam edge curve, extract pixel coordinates of endpoints a and c of the beam edge curve based on the beam segmentation mask, obtain a straight line segment ac, and calculate pixel coordinates of a midpoint b of the straight line segment ac.

[0110] The size measurement unit is configured to calculate a length of the straight line segment ac, convert the length to obtain an actual span of the beam body through the conversion coefficient, and calculate a deflection value.

[0111] The result generating module is configured to output the results of the size measurement unit.

[0112] In a third aspect, based on the same inventive concept, the present application also provides a building structure deformation measurement device, comprising a processor, a memory and a bus, the memory stores instructions and data readable by the processor, the processor is configured to call the instructions and data in the memory to execute the building structure deformation measurement method as described above, and the bus is connected between the functional components for transmitting information.

[0113] In an embodiment, the size measurement device further comprises a laser range finder, a monocular camera and an electric drive platform.​

[0114] In an implementable embodiment, the laser range finder further comprises an angle motor and an angle sensor.

[0115] With the technical scheme, the application has the following beneficial effects:

[0116] The application provides a building structure deformation measurement method, system and device, an electric drive platform supports three-dimensional movement, supports angle adjustment of a laser range finder, and can adjust a laser point positioning position according to the size and position of a building structure; four laser range finders can be adjusted in angle to adapt to building (beam column) structures of different heights and angles, and the limitation that the prior art can only measure specific types or sizes of building structures is solved.

[0117] The scheme establishes a position correction model through three-degree-of-freedom calibration in the early stage, combines multiple measurements and mean filtering processing of laser range reading, and the error of a laser point physical coordinate is less than or equal to 0.1 mm. The system adopts a sub-pixel level precision image processing algorithm to identify a laser spot center, combines camera calibration parameters and a laser projection angle to perform coordinate conversion, and the deflection measurement error is less than or equal to 0.1 mm, and the span measurement error is less than or equal to 0.5%, which is much higher than the precision level of a traditional measurement method.

[0118] The scheme adopts an automatic control mode, can realize automatic operation from laser projection, data acquisition, image processing to result output, four laser range finders work cooperatively, and precise positioning is realized through angle adjustment and distance measurement without manual intervention.

[0119] The scheme adopts a non-contact measurement method, and the laser range finder and the monocular camera do not need to contact the surface of a building structure, measurement can be completed in a safe distance, and a target does not need to be arranged on the surface of the building structure, so that safety risks brought by high-altitude operation or close-range measurement are avoided, and the surface of the building structure is not damaged, and the scheme is particularly suitable for deformation detection of a building structure in use. BRIEF DESCRIPTION OF DRAWINGS

[0120] In order to more clearly illustrate the specific embodiments of the application or the technical scheme in the prior art, the drawings needed in the following description of the specific embodiments or the prior art will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0121] Figure 1 A building structure deformation measurement method flowchart is provided for the embodiments of the application.

[0122] Figure 2It is an example diagram of the main stage in step 3; wherein, (a) is a Gaussian filter image; (b) is an HSV color domain conversion image; (c) is a binary mask image; (d) is a laser point region image;

[0123] Figure 3 It is a schematic diagram of the laser point physical coordinate calculation process in step 4;

[0124] Figure 4 It is a schematic diagram of the relative coordinate calculation process of the object surface laser point in step 4;

[0125] Figure 5 It is a schematic diagram of the projection point provided by the embodiment of the application;

[0126] Figure 6 It is a schematic diagram of the deflection calculation provided by the embodiment of the application. DETAILED DESCRIPTION

[0127] The technical solutions of the present application will be described clearly and completely below with reference to the drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0128] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second", "third" are only for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0129] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0130] The present application will be further explained and described below in conjunction with specific embodiments.

[0131] It should be further explained that the following specific embodiments or specific implementation manners are a series of optimized setting modes listed by the present application for further explaining specific inventive content, and these setting modes can be used in combination or in association with each other.

[0132] Embodiment one:

[0133] As shown in the figure, the building structure deformation measurement method provided by the embodiment comprises the following steps: Figures 1-6 Step 1, set up the size measurement device and perform preliminary calibration, which specifically comprises a laser range finder, a monocular camera and an electric drive platform; the monocular camera is located at the center of the table top of the electric drive platform, and the four laser range finders are located at the four corners of the table top of the electric drive platform; the laser emission direction is perpendicular to the table top of the electric drive platform and parallel to the optical axis of the monocular camera; the laser range finder comprises an angle (for example, pitch and rotation) motor and an angle sensor for controlling the laser emission angle.

[0134] Further, the preliminary calibration specifically comprises:

[0135] Calibrate the monocular camera by Zhang Zhengyou plane calibration method to obtain and store internal and external parameters (such as focal length, pixel size and distortion coefficient, etc.);

[0136] Move the table top of the electric drive platform to the standard calibration area, calibrate the initial physical coordinates of the four laser range finders through the standard size calibration board (a building with known length and width), and record the installation offset of each laser range finder relative to the center of the table top;

[0137] Move the table top of the electric drive platform along the X, Y and Z directions by a preset distance, compare the laser range finder readings with the actual moving distance of the table top, establish a three-degree-of-freedom position correction model, including the current positioning coordinates of the table top, for subsequent coordinate correction;

[0138] Calibrate the angle motor and the angle sensor to establish an angle measurement model, including the angle between the laser emission direction and the world coordinate system.

[0139] Further, the electric drive platform comprises a servo motor and / or a stepping motor and / or a linear module, so as to realize three-dimensional movement and accurate positioning of the table top.

[0140] Step 2, move the table top of the electric drive platform, rotate the angle motor, and make the laser points (light spots) of each laser range finder respectively project to the corresponding corner points of the building facade to be measured, then collect the laser ranging data after preliminary calibration as the physical coordinates of the laser points, collect the angle between the laser emission direction and the world coordinate system through the angle sensor, and collect the building image through the monocular camera at the same time; the building image contains the laser points and the building facade.

[0141] ​

[0142] Furthermore, the physical coordinates of the laser point belong to the world coordinate system and can be denoted as ( ). ), ( ), ( )and( ).

[0143] Step 3: After preprocessing the building image, extract the pixel coordinates of the laser points to provide a high-precision (±1 pixel) image reference for subsequent calculations.

[0144] Furthermore, step 3 specifically includes:

[0145] Step 31: Obtain the candidate region mask for laser points through noise suppression, HSV color gamut conversion, mask extraction and mask morphology optimization.

[0146] Furthermore, step 31 specifically includes:

[0147] Step 311: Apply a Gaussian convolution kernel of a preset size (e.g., 5x5) with the standard deviation of the Gaussian function. (Set to 1.2) The original image of the building image is smoothed to obtain a smooth image. This can effectively suppress Gaussian noise and salt-and-pepper noise by using the normal distribution characteristics and weighted averaging of adjacent pixels.

[0148] Step 312: Select the original image as the guide image and perform guided filtering on the smoothed image (filter radius is 5, regularization parameter is 5). (Setting the value to 100) yields a Gaussian filtered image, such as Figure 2 As shown in Figure (a), this can eliminate high-frequency interference from building surface texture while preserving the grayscale difference between the laser point and the background, thereby ensuring that the edges of the laser point are clear and not blurry.

[0149] Step 313: Convert the Gaussian-filtered image to the HSV color gamut to enhance the distinction between the laser points and the background, such as... Figure 2 As shown in Figure (b); set the V channel threshold range, for example... High-brightness areas are selected from the Gaussian-filtered image;

[0150] In particular, the threshold range of the V channel can be dynamically corrected according to the mean brightness of the Gaussian filtered image ±10, so as to avoid missed detection in strong light or weak light environments.

[0151] Step 314: Set the threshold range for the S channel, for example... Remove pure white background reflections (such as strong reflections on steel structure surfaces) from the Gaussian filtered image.

[0152] Step 315: Perform a logical AND operation on the filtering results of the V and S channels to generate a binary mask containing only the candidate regions of laser points, thereby eliminating more than 95% of non-target pixels and significantly reducing the computational load of subsequent processing. Figure 2 As shown in Figure (c);

[0153] Step 316: Perform morphological closing operation (3*3 rectangular structuring element) on the binarized mask to fill the (tiny) holes inside the laser points (caused by noise); then perform morphological opening operation (3*3 rectangular structuring element) to remove noise attached to the mask edges, and obtain a laser point candidate region mask (with complete outline and clean interior).

[0154] Step 32: Based on the laser point candidate region mask, after extracting the candidate regions, the laser point region image is determined through edge detection and contour feature filtering to avoid interference from workshop light reflections, building surface markers, etc. Figure 2 As shown in Figure (d) of the document.

[0155] Furthermore, step 32 specifically includes:

[0156] Step 321: Extract the edge of the Gaussian-filtered image after the laser point candidate region masking to obtain the candidate region. Then, based on the preset dual thresholds (e.g., the low threshold is 50 and the high threshold is 150), extract the continuous edge contour of the laser point through Gaussian gradient calculation, non-maximum suppression and hysteresis thresholding to ensure that the edge is free of breakage and redundant edge.

[0157] Step 322: (using the findContours function of OpenCV tool) After extracting all contours of the candidate region, set the filtering conditions based on the geometric features of the laser point (circle or approximately circle) to filter out the laser point region image.

[0158] Furthermore, the screening criteria include:

[0159] Filtering criterion 1: Based on the actual size of the laser spot (the pixel area after imaging is usually between 50-200 pixels), set the area threshold range, for example... This is to eliminate excessively small noise contours and excessively large background interference areas;

[0160] Filtering criterion 2: Based on roundness Calculation formula ,in, Represents the area of ​​the outline. This represents the perimeter of the outline, and sets a roundness threshold range, for example... To eliminate irregular non-circular outlines (such as rectangular markers, striped reflections, etc.);

[0161] The screening condition 3 is that a ratio of a minimum circumscribed rectangle of the contour to an area of the contour is calculated, and a contour with a ratio less than or equal to a preset proportion threshold (for example, 1.5) is screened out, so as to further exclude non-laser point interference.

[0162] The step 33 is specifically as follows.

[0163] Further, the step 33 specifically includes:

[0164] The step 331 is that a least square method is used to perform circle fitting on the edge contour of the laser point, so as to solve the center coordinate, and a specific calculation formula includes:

[0165] ;

[0166] Wherein, represents the center coordinate; represents the radius of the circle;

[0167] All pixel points on the edge contour are taken as samples to construct an error function, and the center coordinate and the radius of the circle are solved by taking the partial derivative of the error function to minimize the error function; a specific expression of the error function includes:

[0168] ;

[0169] Wherein, represents the coordinate of the i-th pixel point on the edge contour;

[0170] In particular, if the error function exceeds an error threshold, the laser point region image is re-screened or the preprocessing parameter is optimized to ensure the fitting accuracy; the error threshold is less than or equal to 0.3 pixels.

[0171] The step 332 is that since the building image can be multiple frames (for example, 5 frames) of images under the same shooting condition, and the above steps are performed respectively to obtain multiple groups (for example, 5 groups) of center coordinates, the Grubbs criterion is used to eliminate abnormal center coordinates (i.e., the center coordinates deviating from the mean value by more than 2 times the standard deviation) in the multiple groups of center coordinates derived from the multiple frames of images, and the arithmetic mean of the remaining center coordinates is taken to obtain four (precise) laser point center pixel coordinates, i.e., (x1, y1), (x2, y2), (x3, y3), (x4, y4), (x5, y5), and (x6, y6).

[0172] ​​​Step 4: Based on the physical coordinates of the laser point and the center pixel coordinates of the laser point, calculate the relative coordinates of the laser point on the object surface, solve the homography transformation matrix, perform perspective correction on the building image, obtain the orthophoto image, and obtain the conversion coefficient between the pixel and the actual distance. This eliminates the errors caused by the perspective distortion of the monocular camera and the non-ideal plane of the building surface, establishes a high-precision mapping relationship between the pixel and the actual distance, and provides a benchmark for subsequent measurements.

[0173] Furthermore, step 4 specifically includes:

[0174] Step 41, as Figure 3 As shown, based on previous calibration data and laser ranging data, the physical coordinates of the laser point are solved through spatial geometric calculations. The specific calculation formulas include:

[0175] ;

[0176] ;

[0177] ;

[0178] in, , , Indicates the first Physical coordinates of each laser point; , , Indicates the current positioning coordinates of the platform; , Indicates the first The installation offset of the laser rangefinder relative to the center of the platform; Indicates the first Real-time laser ranging data collected by a laser rangefinder; , , This represents the angle between the laser emission direction and the world coordinate system.

[0179] Step 42: Define the local coordinate system of the object surface. Among them, the laser point in the upper left corner ( The projection point of ) is taken as the origin. , The axis runs horizontally to the right along the building facade (i.e., in relation to the world coordinate system). (axis parallel) The axis runs vertically downwards along the building facade (i.e., in relation to the world coordinate system). (axis parallel) The axis is along the normal vector direction of the building facade (which can be obtained by fitting the physical coordinates of 4 laser points);

[0180] like Figures 4-5The four laser point physical coordinates are projected to the object plane local coordinate system as shown in the formula (1) The relative coordinates of the object plane laser points are obtained in the plane, and the horizontal projection error caused by the axis tilt (small tilt of the building surface) is corrected. The specific expression includes:

[0181] The first projection point , i.e. the origin , has a relative coordinate of ;

[0182] The relative coordinate expression of the second projection point includes:

[0183] ;

[0184] ;

[0185] The relative coordinate expression of the third projection point includes:

[0186] ;

[0187] ;

[0188] The relative coordinate expression of the fourth projection point includes:

[0189] ;

[0190] .

[0191] Step 43, solve the homography transformation matrix, which is used to calculate the corresponding pixel point in the image pixel coordinate system for any point in the object plane local coordinate system, so as to quickly complete the mutual conversion between the two types of coordinates, and provide the basis for subsequent image measurement, target positioning, etc. The specific expression includes:

[0192] ;

[0193] wherein , represents the relative coordinates of the object plane laser points; , represents the pixel coordinates of the laser point center; represents a 3*3 homography matrix, and the specific expression is:

[0194] ;

[0195] wherein represents the matrix elements (there are 9 elements in total, but due to the homogeneity, 8 unknown elements need to be solved)​

[0196] The specific expression unfolded in the non-homogeneous form includes:

[0197] ;

[0198] In this way, by using the known point pairs between the object plane laser point relative coordinates and the laser point center pixel coordinates, a linear equation set is constructed, and then the (4 known point pairs, 8 linear equations are constructed to solve 8 unknown elements), specifically:

[0199] The object plane laser point relative coordinate set is constructed , and the specific expression includes:

[0200] ;

[0201] The laser point center pixel coordinate set is constructed , and the specific expression includes:

[0202] ;

[0203] For each known point pair, two linear equations are obtained by combining the homogeneous coordinate relationship and eliminating the scale factor, for example:

[0204] Substituting the perspective transformation formula into (by fixing the scale through homogeneity, simplifying the calculation), we get:

[0205] ;

[0206] Again, the standard linear equation form is:

[0207] ;

[0208] 4 known point pairs can construct 8 linear equations to form a matrix form:

[0209] ;

[0210] Wherein, represents an 8*8 coefficient matrix, which can be filled by the coordinate values of the 4 known point pairs and according to the above process; represents an 8*1 unknown vector, that is, ; represents an 8*1 constant vector, which can be arranged in corresponding order by the coordinate values of the 4 known point pairs and ;

[0211] Due to the existence of small noise in actual data (e.g. image recognition error, laser point positioning error, etc.), Generally, there is no strict solution, and the optimal solution is obtained by least square method.

[0212] Preferably, the least square method is solved by Singular Value Decomposition (SVD) method, and the specific steps include:

[0213] Step a1, singular value decomposition is performed on to obtain , wherein represents an 8*8 first orthogonal matrix, represents an 8*8 diagonal matrix, represents an 8*8 second orthogonal matrix transpose of

[0214] Step a2, taking the column vector corresponding to the minimum singular value in and , the optimal solution of the linear equation is obtained, that is .

[0215] Step a3, filling the 8 elements in into in order; then (based on the homogeneity assumption) setting =0, =0 and =1, to obtain .

[0216] Step 44, based on the homography transformation matrix, the building image is perspective corrected by setting the corrected image size and inverse perspective transformation (in OpenCV vision library), and the specific expression includes:

[0217] .

[0218] wherein represents the inverse matrix of , which is used to reverse the transformation from image to object plane to orthographic mapping from object plane to image, thereby eliminating perspective distortion; and represent the corrected coordinates of and respectively.

[0219] The corrected image size refers to first calculating the maximum relative coordinates in , which specifically includes:

[0220] horizontal maximum relative coordinates , i.e. the horizontal distribution range covering all laser points, the specific expression is: ;

[0221] Vertical maximum relative coordinate , i.e. the vertical distribution range covering all laser points, the specific expression is: ;

[0222] According to the measurement accuracy requirement (for example, 0.1mm / pixel), the maximum relative coordinate is converted into the number of pixels, which specifically includes:

[0223] Horizontal pixel number , the specific expression is: , wherein, represents the rounding operation;

[0224] Vertical pixel number , the specific expression is: ;

[0225] In this way, the image size after correction can be matched with the laser point distribution of the local coordinate system of the object surface, so as to ensure complete coverage of the object surface area and reasonable pixel accuracy;

[0226] In particular, if * is lower than 500*500 pixels, the accuracy requirement can be reduced (for example, adjusting 0.1mm / pixel to 0.2mm / pixel) and then recalculated to ensure that the image effective information is complete and details are not lost;

[0227] The inverse perspective transformation can be realized by the warpPerspective function, which specifically includes: taking the preprocessed building image (i.e. the Gaussian filtered image obtained in step 312) as the input image; taking as the transformation matrix; taking * as the output size; taking the bicubic interpolation algorithm as the interpolation algorithm; calling the warpPerspective function to obtain the orthographic projection image (without perspective distortion);

[0228] Step 45, calculate the conversion coefficient between the pixel and the actual distance, the specific expression includes:

[0229] .

[0230] Preferably, the step 45 specifically includes:

[0231] Step 451, calculate the actual distance mean value of the horizontal laser point and the pixel distance mean value , the specific expression includes:

[0232] ;

[0233] ;

[0234] Calculate the average of the actual distance of the vertical laser point and the average of the pixel distance , the specific expression includes:

[0235] ;

[0236] ;

[0237] Step 452, calculate the conversion coefficient between pixels and actual distance , unit: mm / pixel, the specific expression includes:

[0238] ;

[0239] In this way, when the pixel distance in each direction is known, the actual distance can be quickly obtained by directly multiplying , and vice versa.

[0240] Step 5, input the preprocessed building image (i.e. the Gaussian filtered image obtained in step 312) into the pre-trained semantic segmentation model to obtain the beam body segmentation mask and the beam body edge curve; based on the beam body segmentation mask, the pixel coordinates of the two endpoints a and c of the beam body edge curve are extracted, and the straight line segment ac is obtained, then the pixel coordinates of the midpoint b of the straight line segment ac are calculated, so as to obtain the accurate positions of the three feature points, and the automatic recognition and feature point extraction of the beam body region are completed.

[0241] Further, the specific training method of the semantic segmentation model includes: after pre-processing different sizes and different angles of building images, labeling the beam body region and the beam body edge curve, generating a pixel-level mask label as a training data set; training the semantic segmentation model (such as deeplabv3 deep learning model, U-Net semantic segmentation model, and SegNet semantic segmentation model, etc.), the input is the preprocessed building image, and the output is the mask label; through back propagation optimization model parameters, so that it has the ability to accurately segment the beam body region and extract the beam body edge curve.

[0242] Step 6, as shown in Figure 6 , the length of the straight line segment ac is calculated, and the actual span of the beam body is converted through the conversion coefficient; similarly, the height difference between the endpoint a and the midpoint b, the height difference between the endpoint a and the endpoint c, and the horizontal distance between the endpoint a and the midpoint b are converted.and the horizontal distance between the midpoint b and the endpoint c Then, the deflection value is calculated The specific formula is:

[0243] ;

[0244] In this way, the actual span and the deflection value are combined with the building type query to obtain the corresponding building specification, so that the compliance judgment can be accurately performed.

[0245] Further, the conversion in step 6 refers to the pixel distance of the measured straight line segment multiplied by the conversion coefficient.

[0246] Embodiment two:

[0247] The embodiment provides a building structure deformation measurement system, comprising a data receiving module, a data processing module and a result generating module.

[0248] The data receiving module is configured to receive laser ranging data as physical coordinates of laser points, an angle between a laser emission direction and a world coordinate system, and a building image.

[0249] The data processing module comprises a preprocessing unit, a perspective correction unit, a semantic segmentation unit and a size measurement unit.

[0250] The preprocessing unit is configured to preprocess the building image and extract pixel coordinates of the laser points.

[0251] The perspective correction unit calculates the relative coordinates of the object surface laser points based on the physical coordinates of the laser points and the pixel coordinates of the laser point centers, solves a homographic transformation matrix, corrects the perspective of the building image, obtains an orthographic projection image and obtains a conversion coefficient between the pixel and the actual distance.

[0252] The semantic segmentation unit is configured to input the preprocessed building image into a pre-trained semantic segmentation model to obtain a beam body segmentation mask and a beam body edge curve; based on the beam body segmentation mask, the pixel coordinates of the two endpoints a and c of the beam body edge curve are extracted to obtain a straight line segment ac, and the pixel coordinates of the midpoint b of the straight line segment ac are calculated.

[0253] The size measurement unit calculates the length of the straight line segment ac, converts the actual span of the beam body through the conversion coefficient, and calculates the deflection value ;

[0254] The result generating module is configured to output the result of the size measurement unit.

[0255] Embodiment three:

[0256] This embodiment provides a building structure deformation measurement device, including a processor, a memory, and a bus. The memory stores instructions and data that can be read by the processor. The processor is used to call the instructions and data in the memory to execute the building structure deformation measurement method described above. The bus connects the various functional components for transmitting information.

[0257] Furthermore, the size measuring device also includes a laser rangefinder, a monocular camera, and an electric drive platform.

[0258] Furthermore, the laser rangefinder also includes an angle motor and an angle sensor.

[0259] In another embodiment, this solution can also be implemented using an integrated device, which may include corresponding modules that perform one or more steps in the various embodiments described above. A module may be one or more hardware modules specifically configured to perform the corresponding step, or implemented by a processor configured to perform the corresponding step, or stored in a computer-readable medium for implementation by a processor, or implemented through some combination thereof.

[0260] The processor executes the various methods and processes described above. For example, the method implementations in this scheme can be implemented as software programs tangibly contained in a machine-readable medium, such as memory. In some implementations, part or all of the software program can be loaded and / or installed via memory and / or a communication interface. When the software program is loaded into memory and executed by the processor, one or more steps of the methods described above can be performed. Alternatively, in other implementations, the processor can be configured to execute one of the methods described above by any other suitable means (e.g., by means of firmware).

[0261] This device can be implemented using a bus architecture. A bus architecture can include any number of interconnect buses and bridges, depending on the specific application of the hardware and overall design constraints. The bus connects various circuits, including one or more processors, memory, and / or hardware modules. The bus can also connect various other circuits such as peripherals, voltage regulators, power management circuitry, external antennas, etc.

[0262] Buses can be Industry Standard Architecture (ISA) buses, Peripheral Component Interconnect (PCI) buses, or Extended Industry Standard Component (EISA) buses, etc. Buses can be divided into address buses, data buses, control buses, etc.

[0263] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions recorded in the above embodiments can be modified, or some or all of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for measuring structural deformation of a building, characterized in that, include: Step 1: Set up the size measurement device and perform preliminary calibration, specifically including a laser rangefinder, a monocular camera, and an electric drive platform; the monocular camera is located at the center of the electric drive platform, and four laser rangefinders are located at the four corners of the electric drive platform; the laser emission direction is perpendicular to the electric drive platform and parallel to the optical axis of the monocular camera; the laser rangefinder includes an angle motor and an angle sensor, used to control the laser emission angle; Step 2: Move the platform of the electric drive platform, rotate the angle motor so that the laser points of each laser rangefinder are projected onto the corresponding corner points of the facade of the building to be measured. Collect the laser ranging data after the previous calibration as the physical coordinates of the laser points. Collect the angle between the laser emission direction and the world coordinate system through the angle sensor. At the same time, collect the building image through the monocular camera. Step 3: After preprocessing the building image, extract the pixel coordinates of the laser points; Step 4: Based on the physical coordinates of the laser point and the center pixel coordinates of the laser point, calculate the relative coordinates of the laser point on the object surface, solve the homography transformation matrix, perform perspective correction on the building image, obtain the orthographic projection image, and obtain the conversion coefficient between the pixel and the actual distance. Step 5: Input the preprocessed building image into the pre-trained semantic segmentation model to obtain the beam segmentation mask and beam edge curve; based on the beam segmentation mask, extract the pixel coordinates of the two endpoints a and c of the beam edge curve, obtain the straight line segment ac, and then calculate the pixel coordinates of the midpoint b of the straight line segment ac. Step 6: Calculate the length of straight segment ac, and use the conversion factor to obtain the actual span of the beam; similarly, calculate the height difference between endpoint a and midpoint b. The height difference between endpoint a and endpoint c Horizontal distance between endpoint a and midpoint b and the horizontal distance between midpoint b and endpoint c Then, calculate the deflection value. The specific formula is as follows: 。 2. The measurement method according to claim 1, characterized in that, The preliminary calibration specifically includes: The monocular camera was calibrated using Zhang Zhengyou's flat-panel calibration method, and its internal and external parameters were acquired and stored. Move the electric drive platform to the standard calibration area, calibrate the initial physical coordinates of the four laser rangefinders using a standard-size calibration plate, and record the installation offset of each laser rangefinder relative to the center of the platform. The electric drive platform is moved a preset distance along the X, Y and Z directions respectively. The laser rangefinder readings are compared with the actual distance the platform moves to establish a three-degree-of-freedom position correction model, including the current positioning coordinates of the platform. The angle motor and angle sensor are calibrated, and an angle measurement model is established, including the angle between the laser emission direction and the world coordinate system.

3. The measurement method according to claim 1, characterized in that, Step 3 specifically includes: Step 31: Obtain the candidate region mask for laser points through noise suppression, HSV color gamut conversion, mask extraction and mask morphology optimization; Step 32: Based on the candidate region mask for laser points, after extracting the candidate regions, determine the laser point region image through edge detection and contour feature filtering; Step 33: For the laser point area image, the center pixel coordinates are solved by fitting algorithm, and the final center pixel coordinates are obtained by multi-frame averaging optimization.

4. The measurement method according to claim 3, characterized in that, Step 31 specifically includes: Step 311: Smooth the original image of the building image by using a Gaussian convolution kernel of a preset size to obtain a smoothed image; Step 312: Select the original image as the guide image, perform guided filtering on the smoothed image, and obtain a Gaussian filtered image; Step 313: Convert the Gaussian filtered image to the HSV color gamut, set the threshold range for the V channel, and filter out the high-brightness areas from the Gaussian filtered image; Step 314: Set the threshold range for the S channel to remove pure white background reflections from the Gaussian filtered image; Step 315: Perform a logical AND operation on the filtering results of the V channel and S channel to generate a binary mask that only contains the candidate regions of laser points; Step 316: Perform morphological closing operation on the binarized mask to fill the holes inside the laser points; then perform morphological opening operation to remove noise attached to the edge of the mask, and obtain the candidate region mask for the laser points.

5. The measurement method according to claim 4, characterized in that, Step 32 specifically includes: Step 321: Extract the edge of the Gaussian filtered image after the laser point candidate region masking to obtain the candidate region. Then, based on the preset double threshold, extract the continuous edge contour of the laser point through Gaussian gradient calculation, non-maximum suppression and hysteresis thresholding. Step 322: After extracting all contours of the candidate region, set filtering conditions based on the geometric features of the laser points to filter out the laser point region images.

6. The measurement method according to claim 5, characterized in that, The filtering criteria include: Filtering condition 1: Set the area threshold range based on the actual size of the laser point; Filtering criterion 2: Based on roundness Calculation formula ,in, Represents the area of ​​the outline. Indicates the perimeter of the outline, and sets the roundness threshold range; Filtering condition 3: Calculate the ratio of the minimum bounding rectangle of the contour to the area of ​​the contour, and filter out contours with a ratio ≤ a preset ratio threshold.

7. The measurement method according to claim 3, characterized in that, Step 33 specifically includes: Step 331: Using the least squares method, fit a circle to the edge contour of the laser point and solve for the coordinates of the circle center. The specific calculation formula includes: ； in, Indicates the coordinates of the center of the circle; Indicates the radius of the circle; Using all pixels on the edge contour as samples, an error function is constructed. By minimizing the error function through partial derivatives, the coordinates of the circle center and the circle radius are obtained. The specific expressions include: ； in, Indicates the first edge contour The coordinates of each pixel; Step 332: The architectural images are multiple frames taken under the same conditions. The above steps are performed on each frame to obtain multiple sets of center coordinates. Using the Grubbs criterion, abnormal center coordinates from the multiple sets of center coordinates from the multiple frames are removed. The arithmetic mean of the remaining center coordinates is taken to obtain the center pixel coordinates of the four laser points. ( ), ( ), ( )and ( ).

8. The measurement method according to claim 2, characterized in that, Step 4 specifically includes: Step 41: Based on the previous calibration data and laser ranging data, solve for the physical coordinates of the laser point through spatial geometric calculations. The specific calculation formulas include: ； ； ； in, , , Indicates the first Physical coordinates of each laser point; , , Indicates the current positioning coordinates of the platform; , Indicates the first The installation offset of the laser rangefinder relative to the center of the platform; Indicates the first Real-time laser ranging data collected by a laser rangefinder; , , Indicates the angle between the laser emission direction and the world coordinate system; Step 42: Define the local coordinate system of the object surface. The origin is defined by the projection of the laser point in the upper left corner. , The axis runs horizontally to the right along the building facade; The axis runs vertically downwards along the building facade; The axis is along the direction of the normal vector of the building facade; Project the physical coordinates of the four laser points onto the local coordinate system of the object surface. Within the plane, the relative coordinates of the laser point on the object surface are obtained and corrected. The horizontal projection error caused by axis tilt is specifically expressed as follows: First projection point The origin The relative coordinates are ; Second projection point The relative coordinate expressions include: ； ； Third projection point The relative coordinate expressions include: ； ； Fourth projection point The relative coordinate expressions include: ； ； Step 43: Solve for the homography transformation matrix. The specific expression includes: ； in, , Indicates the relative coordinates of the laser point on the object surface; , Indicates the center pixel coordinates of the laser point; Represents a 3x3 homography matrix; Step 44: Based on the homography transformation matrix, perform perspective correction on the architectural image by setting the corrected image size and inverse perspective transformation. The specific expressions include: ； in, express The inverse matrix; and They represent and Corrected coordinates; The corrected image size refers to the size of the image first... In the calculation of the maximum relative coordinates, the specific steps include: Horizontal maximum relative coordinates The specific expression is: ; Vertical maximum relative coordinates The specific expression is: ; Then, based on the required measurement accuracy, the maximum relative coordinates are converted into the number of pixels, specifically including: Horizontal pixels The specific expression is: ,in, This indicates the rounding operation; Vertical pixel count The specific expression is: ; Step 45: Calculate the conversion coefficient between pixels and actual distance. The specific expression includes: 。 9. A building structure deformation measurement system employing the measurement method described in any one of claims 1-8, characterized in that, It includes a data receiving module, a data processing module, and a result generation module; The data receiving module is used to receive laser ranging data as the physical coordinates of the laser point, the angle between the laser emission direction and the world coordinate system, and building images. The data processing module includes a preprocessing unit, a perspective correction unit, a semantic segmentation unit, and a size measurement unit. The preprocessing unit is used to preprocess the building image and extract the pixel coordinates of the laser points; The perspective correction unit calculates the relative coordinates of the laser point on the object surface based on the physical coordinates of the laser point and the center pixel coordinates of the laser point, solves the homography transformation matrix, performs perspective correction on the building image, obtains the orthographic projection image, and obtains the conversion coefficient between the pixel and the actual distance. The semantic segmentation unit is used to input the preprocessed building image into the pre-trained semantic segmentation model to obtain the beam segmentation mask and the beam edge curve; based on the beam segmentation mask, the pixel coordinates of the two endpoints a and c of the beam edge curve are extracted to obtain the straight line segment ac, and then the pixel coordinates of the midpoint b of the straight line segment ac are calculated. The dimensional measurement unit calculates the length of the straight segment ac, and converts it into the actual span of the beam using a conversion factor. Calculate deflection value ; The result generation module is used to send the results of the size measurement unit to an external source.

10. A device for measuring structural deformation of a building, characterized in that, It includes a processor, a memory, and a bus. The memory stores instructions and data read by the processor. The processor is used to call the instructions and data in the memory to execute the measurement method as described in any one of claims 1-8. The bus connects the functional components for transmitting information.

Images