Calibration-free large structure deformation laser vision fusion measurement method and system

Abstract

The invention discloses a calibration-free large-scale structure deformation laser vision fusion measurement method and system, and the method comprises the steps: setting a visual image collection instrument, enabling a laser transmitter to transmit laser rays on the inner wall of a target structure, and enabling the laser rays to be imaged on the visual image collection instrument; and determining a target offset of a pixel point between the laser line and the reference line based on a target image formed on the acquisition instrument. And before this, fitting the corresponding relation between the pixel displacement and the proportionality coefficient by controlling the actual displacement of the measured object and the pixel displacement corresponding to the imaging of the visual image acquisition instrument in the displacement process to obtain a predetermined relation table. After a target offset corresponding to the target image is obtained, obtaining a proportionality coefficient corresponding to the target offset by querying a predetermined relation table; and converting the target offset into an actual physical displacement generated by deformation of the target structure based on the proportionality coefficient. The actual displacement can be obtained only through the pixel displacement, and calibration does not need to be carried out in an on-site complex working condition environment.

Description

Technical Field

[0001] This application relates to the technical field of deformation monitoring, and particularly to a method and system for laser-vision fusion measurement of deformation of large-scale structures without calibration. Background Art

[0002] Currently, for the monitoring of the structural deformation of various structures in infrastructure, methods such as laser ranging method and three-dimensional laser scanning method are usually used. However, in the application process of the above methods, external parameter calibration is often required before use under various complex working conditions, which has certain calibration difficulties and complexities. Summary of the Invention

[0003] An embodiment of this application provides a method and system for laser-vision fusion measurement of deformation of large-scale structures without calibration to solve the technical problem of difficult calibration under complex working conditions.

[0004] According to the first aspect of the embodiment of this application, a method for laser-vision fusion measurement of deformation of large-scale structures without calibration is provided. The method includes: collecting a target image formed by a laser line emitted by a laser emitter in a target structure through a vision image collector; determining a target offset of pixel points between the laser line and a reference line based on the target image; obtaining a proportional coefficient corresponding to the target offset by querying a predetermined relationship table; the predetermined relationship table records the corresponding relationship between the offset and the proportional coefficient; converting the target offset into a physical displacement based on the proportional coefficient; and determining deformation data of the target structure according to the physical displacements corresponding to all pixel points.

[0005] Further, before collecting a target image formed by a laser line emitted by a laser emitter in a target structure through a vision image collector, the method further includes: collecting a measurement image formed by the laser line emitted by the laser emitter during a predetermined number of movements of a measurement object through the vision image collector; fitting based on the number of movements of the measurement object, the actual displacement of the measurement object, and the pixel offset in the measurement image to obtain a fitting function between the proportional coefficient and the pixel offset; determining the corresponding relationship between the proportional coefficient and the pixel offset based on the fitting function; and recording the corresponding relationship in a predetermined relationship table.

[0006] Further, the fitting based on the number of movements of the measurement object, the actual displacement of the measurement object, and the pixel offset in the measurement image includes: fitting based on the number of movements of the measurement object, the actual displacement of the measurement object in each movement, the proportional coefficient corresponding to the change in the angle between the laser plane and the imaging plane caused by each movement, and the pixel offset in the measurement image.

[0007] Further, there are multiple measurement objects, and the moving direction of the measurement objects is consistent with the deformation direction along the radial direction of the target structure.

[0008] Further, determining the corresponding relationship between the proportional coefficient and the pixel offset based on the fitting function includes: constructing the Euclidean norm of the residual vector in the fitting function based on the least squares method; taking the derivative of the Euclidean norm, and determining the corresponding relationship between the proportional coefficient and the pixel offset when the result of the derivative is equal to 0.

[0009] Further, determining the target offset of the pixel points between the laser line and the reference line based on the target image includes: after denoising the target image, determining whether the background interference complexity in the target image reaches a predetermined condition; in response to the background interference complexity reaching the predetermined condition, adjusting the threshold corresponding to the target image based on the local window mean, standard deviation, and empirical coefficient corresponding to the target image; determining the pixel coordinates of the laser line in the target image based on a predetermined algorithm and the threshold; determining the target offset of the pixel points between the laser line and the reference line based on the pixel coordinates.

[0010] Further, the proportional coefficient is related to the scale factor of the visual image acquisition device and the cosine value of the angle between the laser plane and the imaging plane; wherein, the scale factor is equal to the product of the pixel size and the object distance divided by the focal length.

[0011] Further, converting the target offset into a physical displacement amount based on the proportional coefficient includes: multiplying the target offset by the proportional coefficient to obtain the physical displacement amount.

[0012] According to the second aspect of the embodiments of the present application, there is also provided a large-scale structure deformation laser vision fusion measurement system without calibration. The system applies the large-scale structure deformation laser vision fusion measurement method without calibration according to any one of the foregoing first aspects. The system includes: a processing device, and at least a pair of visual image acquisition devices and laser emitters; the at least a pair of visual image acquisition devices and laser emitters are arranged on the inner wall of at least one section of the target structure, and the distance between a pair of visual image acquisition devices and laser emitters is greater than a preset distance; the axis of the laser emitter is parallel to the axis of the target structure, and the processing device is at least communicatively connected to the visual image acquisition device.

[0013] Further, the visual image acquisition device includes: a main camera and an auxiliary camera; the angle between the axis of the main camera and the axis of the auxiliary camera is less than a preset threshold.

[0014] A laser vision fusion measurement method for deformation of large-scale structures without calibration proposed in an embodiment of the present application collects a target image formed by a laser line emitted by a laser emitter in a target structure through a vision image collector; determines a target offset of pixel points between the laser line and a reference line based on the target image; obtains a proportionality coefficient corresponding to the target offset by querying a predetermined relationship table; the predetermined relationship table records the corresponding relationship between the target offset and the proportionality coefficient; converts the target offset into a physical displacement based on the proportionality coefficient; and determines deformation data of the target structure according to the physical displacements corresponding to all pixel points. In this way, during the actual detection process, only by querying the relationship table between the pre-set proportionality coefficient and the target offset, the proportionality coefficient corresponding to the existing offset of the current laser line pixel points can be queried, and then the offset can be converted into the physical displacement actually generated by the target structure itself based on the proportionality coefficient. In this way, the conversion from the laser line offset to the actual deformation of the structure can be achieved only by querying the predetermined relationship table, without the need to calibrate the parameters of the device before each detection, avoiding the low efficiency caused by the calibration difficulty under complex working conditions, and improving the detection efficiency and accuracy. Description of the Drawings

[0015] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings: Figure 1 is a flowchart of a laser vision fusion measurement method for deformation of large-scale structures without calibration according to an embodiment of the present application; Figure 2 is a flowchart of a laser vision fusion measurement method for deformation of large-scale structures without calibration according to an embodiment of the present application; Figure 3 is a schematic diagram of an imaging relationship according to an embodiment of the present application; Figure 4 is a schematic diagram of the angle between a laser plane and an imaging plane according to an embodiment of the present application Figure 1 ; Figure 5 is a schematic diagram of the angle between a laser plane and an imaging plane according to an embodiment of the present application Figure 2 ; Figure 6 is a schematic diagram of the movement of a measurement object according to an embodiment of the present application; Figure 7 is a schematic diagram of a laser vision fusion measurement system for deformation of large-scale structures without calibration according to an embodiment of the present application. Detailed Embodiments

[0016] In order to enable those skilled in the art to better understand the solution of this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application. It should be noted that, without conflict, the embodiments in this application and the features in the embodiments can be combined with each other.

[0017] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of this application described here can be implemented in an order different from those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0018] In this embodiment, a laser vision fusion measurement method for deformation of large-scale structures without calibration is provided. Among them, as Figure 1 shown, by setting a vision image collector, a laser emitter emits a laser line on the inner wall of the target structure and forms an image on the vision image collector, and the target offset of the pixel points between the laser line and the reference line is determined based on the target image formed on the collector. Before that, by controlling the actual displacement of the measured object and the pixel displacement corresponding to the imaging of the displacement process on the vision image collector, the corresponding relationship between the pixel displacement and the proportional coefficient is fitted to obtain a predetermined relationship table. After obtaining the target offset corresponding to the target image, the proportional coefficient corresponding to the target offset is obtained by querying the predetermined relationship table, and the target offset is converted into the actual physical displacement generated by the deformation of the target structure based on the proportional coefficient.

[0019] Figure 2 is a flowchart of a laser vision fusion measurement method for deformation of large-scale structures without calibration according to an embodiment of this application. As Figure 2 shown, the process includes the following steps: S10: Collect the target image formed by the laser emitter emitting a laser line in the target structure through the vision image collector; S20: Determine the target offset of the pixel points between the laser line and the reference line based on the target image; S30: Obtain the proportionality coefficient corresponding to the target offset by querying the pre - defined relationship table; the pre - defined relationship table records the corresponding relationship between the target offset and the proportionality coefficient; S40: Convert the target offset into a physical displacement based on the proportionality coefficient; S50: Determine the deformation data of the target structure according to the physical displacements corresponding to all pixel points.

[0020] In this embodiment, the visual image acquisition device may include a camera. The visual image acquisition device and the laser emitter may exist in pairs. Each visual image acquisition device may be used to acquire the target image formed by a laser emitter emitting a laser line in the target structure. The laser emitter may also be referred to as a laser, and the visual image acquisition device and the laser emitter may be arranged on the inner wall of the target structure. The laser emitter may emit a fan - shaped laser line along the cross - section of the target structure towards the inner wall of the target structure. Here, the cross - section may be a cross - section perpendicular to the radial direction of the target structure. The target structure may be a tunnel or the like.

[0021] In one embodiment, the deformation detection may be based on the principle of laser triangulation to measure the structural deformation of the measured plane. Based on the proportionality coefficient, a geometric relationship model between the laser plane and the imaging plane can be established to eliminate the measurement error caused by the non - parallelism of the two planes. The laser line may also be referred to as the monitoring line. The relationship between the laser plane and the imaging plane is as Figure 3 shown, where O W is the origin of the laser - plane coordinate system, X W , Y W and Z W are the X - axis, Y - axis and Z - axis of the laser - plane coordinate system respectively; O C is the origin of the imaging - plane coordinate system, X C , Y C and Z C are the X - axis, Y - axis and Z - axis of the laser - plane coordinate system respectively. a1, b1, c1 are the pixel positions of the reference line on the laser plane, a2, b2, c2 are the pixel positions of the laser line on the laser plane, a1’, b1’, c1’ are the pixel positions of the reference line on the imaging plane, and a2’, b2’, c2’ are the pixel positions of the laser line on the imaging plane.

[0022] Among them, as Figure 4As shown in the figure, the fan-shaped laser line emitted by the laser emitter forms a laser plane in space, and there is a fixed angle θ between the laser plane and the imaging plane of the visual image acquisition device. When the surface of the object undergoes displacement, the position of the laser line on the visual image acquisition device will shift, that is, the angle change Δθ is generated. By establishing the sine theorem model, under the conditions of knowing the distance between the laser plane and the visual image acquisition device, that is, the object distance L, and the focal length f of the visual image acquisition device, the offset (Δh) of the pixel point can be converted into the actual displacement (Δy). Among them, is the linear calibration coefficient. The formula is obtained: Δh = k1·Δy.

[0023] In one embodiment, the target offset is the offset of the aforementioned pixel point. Determining the target offset of the pixel point between the laser line and the reference line may refer to determining the target offset between the coordinates of each pixel point on the laser line and the coordinates on the reference line.

[0024] In one embodiment, the physical displacement is the aforementioned actual displacement, that is, the displacement of the point position caused by the actual deformation of the target structure. The predetermined relationship table can record the proportionality coefficients corresponding to different target offsets. For example, the predetermined relationship table can be generated based on the fitting function between the target offset and the proportionality coefficient.

[0025] In one embodiment, based on the baseline distance L from the laser emitter to the optical center of the visual image acquisition device, the focal length f, the initial angle α between the normal vector of the laser plane and the reference line, and based on the actual angle θ between the laser plane and the imaging plane as the dynamic calibration parameter, a complete geometric relationship model is established. According to the modified laser point projection model, as Figure 5 shown, when the angle between the laser plane and the imaging plane is θ, the relationship between the pixel displacement Δh (that is, the target offset or pixel offset) and the actual displacement Δy (that is, the physical displacement or actual displacement) can be expressed by the formula:

[0026] By introducing the dynamic compensation mechanism of the angle θ, the measurement accuracy under complex installation conditions is significantly improved.

[0027] In one embodiment, determining the deformation data of the target structure according to the physical displacements corresponding to all pixel points may include: traversing the physical displacements corresponding to all pixel points on the laser line in the target image, and generating a full-field displacement distribution map representing the deformation of the target structure.

[0028] In one embodiment, before step S10, the method may further include: establishing a reference line. Establishing the reference line may include: when the target structure is not deformed or the degree of disturbance is lower than a predetermined value, capturing multiple reference images corresponding to the light emitted by the laser emitter and taking the average value; after eliminating noise interference, extracting the position of the laser line through the Steger algorithm to obtain a stable reference line and saving the position information of each pixel point on the reference line.

[0029] Here, the position information may include coordinate data. Step S20 may include: performing preprocessing such as noise reduction and fracture repair on the target image; extracting the exact position of the current laser line based on the Steger algorithm; comparing the position of each pixel point of the laser line with the position of the corresponding pixel point of the reference line one by one to calculate the target offset.

[0030] Here, the offset may be a lateral offset, that is, the offset in the X-axis direction of the laser plane. Among them, the X-axis and Y-axis of the laser plane are perpendicular to the radial direction of the target structure, and the Z-axis is consistent with the radial direction.

[0031] In this way, during the actual detection process, only by querying the relationship table between the pre-set scale factor and the target offset, the scale factor corresponding to the offset existing in the current laser line pixel points can be queried, and then the offset can be converted into the actual physical displacement amount generated by the target structure itself based on the scale factor. In this way, the conversion from the laser line offset to the actual deformation amount of the structure can be achieved only by querying the predetermined relationship table, without the need to calibrate the parameters of the device before each detection, avoiding the inefficiency caused by the calibration difficulty under complex working conditions and improving the detection efficiency and accuracy.

[0032] In some embodiments, before step S10, the method further includes: Collecting measurement images formed during the movement of the laser line emitted by the laser emitter based on the predetermined number of movements of the measurement object through a visual image acquisition device; Performing fitting based on the number of movements of the measurement object, the actual displacement amount of the measurement object, and the pixel offset in the measurement image to obtain a fitting function between the scale factor and the pixel offset; Determining the corresponding relationship between the scale factor and the pixel offset based on the fitting function; Recording the corresponding relationship in a predetermined relationship table.

[0033] In one embodiment, there are multiple measurement objects, and the moving direction of the measurement objects is consistent with the deformation direction along the radial direction of the target structure. Among them, multiple Invar standard blocks with good thermal stability can be used as the measurement objects, where the measurement objects are used as calibration blocks. The calibration blocks are fixed at points manually, and the moving process can refer to vertically moving the calibration blocks along the radial direction of the target structure. After each movement, the visual image acquisition device is triggered synchronously to collect measurement images, and the actual displacement and the pixel offset of the laser line in the measurement image are recorded.

[0034] Here, the predetermined number of times can be 50 times. Then, a total of 150 groups of data are generated for the three calibration blocks, covering the full range of 0 - 4096 pixels of the cross-section (i.e., the transverse section) of the target structure. Each time the measurement object is moved, there will be a corresponding actual displacement in the measurement image for the actual structure being measured.

[0035] In one embodiment, the fitting based on the number of movements of the measurement object, the actual displacement of the measurement object, and the pixel offset in the measurement image may include: Fitting based on the number of movements of the measurement object, the actual displacement of the measurement object in each movement, the proportional coefficient corresponding to the change in the angle between the laser plane and the imaging plane caused by each movement, and the pixel offset in the measurement image.

[0036] Here, the abscissa of each pixel The corresponding proportional coefficient Can be directly calculated from the relationship between the actual displacement and the pixel offset, that is, based on the number of movements of the measurement object, the actual displacement of the measurement object in each movement, the proportional coefficient corresponding to the change in the angle between the laser plane and the imaging plane caused by each movement And the pixel offset in the measurement image for fitting to obtain a fitting function:

[0037] Where j is the number of movements of the measurement object, Represents the displacement of the pixel, that is, the pixel offset, at the j - th movement. Due to the inclination of the laser plane, the angle Between the laser plane and the imaging plane changes, and thus Changes with the spatial position. For the same abscissa Multiple sets of Are generated in different numbers of movements, which constitute the input data of an over - determined system of equations for fitting. As

[0038] Shown in Figure 6 , for the measurement object at the pixel point position with abscissa x i , the actual displacement of each movement is h, and the proportional coefficient ​It can represent the proportional relationship between the target offset and the physical displacement, or the proportional relationship between the pixel offset and the actual displacement. i is a positive integer greater than 0 and less than or equal to n, and n is the total number of pixel points on the laser line.

[0039] In one embodiment, if the predetermined number of times is 50 times, an observation equation containing 50 sets of movement data is established for each pixel point, and a total of 204,800 equations are generated for 4,096 pixels in the full screen. The fitting function can be further expressed as , where f is a fifth-order polynomial fitting formula.

[0040] In this way, by measuring the multiple movement processes of the object along the deformation direction and collecting the corresponding measurement images, the proportional relationship between the displacement of the pixel points in the measurement image and the actual displacement of the object can be fitted, and the corresponding relationship between the proportional coefficient and the pixel offset can be accurately obtained, which is beneficial to quickly and accurately obtain the required proportional coefficient in the actual deformation detection process.

[0041] In some embodiments, determining the corresponding relationship between the proportional coefficient and the pixel offset based on the fitting function may include: Constructing the Euclidean norm of the residual vector in the fitting function based on the least squares method; Taking the derivative of the Euclidean norm and determining the corresponding relationship between the proportional coefficient and the pixel offset based on the result of the derivative being equal to 0.

[0042] Here, assume there is a set of observed data points: ( ), ( ), …, ( ), ,

[0043] Then the polynomial fitting model, that is, the aforementioned fitting function, can be expressed as:

[0044] where S is a column vector (i.e., the dependent variable) composed of the proportional coefficients corresponding to the measurement image, X is a matrix composed of pixel offsets (i.e., the independent variable), for example, a Vandermonde matrix, A is a coefficient vector to be determined, is the residual vector.

[0045] In one embodiment, constructing the Euclidean norm of the residual vector in the fitting function based on the least squares method can be expressed as: J(A)=||S - XA|| 2 =(S - XA) T (S - XA).

[0046] Further, J(A) = S T S - 2A T X T S + A T X T XA。

[0047] In one embodiment, taking the derivative of the Euclidean norm and determining the correspondence between the proportionality coefficient and the pixel offset based on the result of the derivative being equal to 0 may include: taking the first derivative of the Euclidean norm and setting it to zero to find the minimum value, that is:

[0048] From this, we get:

[0049] When is a full-rank matrix, the normal equations have a unique solution:

[0050] If there exists an ill-conditioned matrix or a non-full-rank situation, the pseudo-inverse can be used:

[0051] Then the predicted value of S after fitting is:

[0052] Therefore, for the complete fifth-order polynomial least squares fitting problem, which is the proportionality coefficient required based on the pixel offset or the target offset, the correspondence between the proportionality coefficient and the pixel offset can be strictly represented in matrix form as: , , .

[0053] In this way, based on the Euclidean norm of the residual vector in the fitting function, that is, the sum of squared errors, and combined with the fact that the derivative being equal to 0 can characterize the minimum error, the correspondence between the proportionality coefficient and the pixel offset can be obtained more accurately, that is, the correspondence between the proportionality coefficient and the target offset.

[0054] In some embodiments, determining the target offset of the pixel points between the laser line and the reference line based on the target image includes: After denoising the target image, determining whether the background interference complexity in the target image reaches a predetermined condition; In response to the background interference complexity reaching a predetermined condition, adjust the threshold corresponding to the target image based on the local window mean, standard deviation, and empirical coefficient of the target image; Determine the pixel coordinates of the laser line in the target image based on a predetermined algorithm and the threshold; Determine the target offset of the pixels between the laser line and the reference line based on the pixel coordinates.

[0055] In one embodiment, in the numerical calculation module, through a high-precision image processing and dynamic compensation algorithm, convert the laser line pixel coordinates into actual deformation data. Based on the laser light strip image collected by the industrial camera, first, Gaussian filtering and morphological closing operations are required to eliminate ambient light interference and speckle noise. After denoising, the sub-pixel coordinates of the center of the laser light strip are extracted through the Steger algorithm. The Steger algorithm calculates the normal direction of the light strip based on the Hessian matrix, and performs gray center of gravity iteration along the normal direction to achieve a light strip center positioning accuracy of ±0.05 pixels. For complex background interference, adaptive threshold segmentation can be used, and its threshold calculation formula is: T(x, y) = μ(x, y) + k * σ(x, y).

[0056] Where μ(x, y) is the local window mean, σ(x, y) is the standard deviation, and the empirical coefficient k can be 1.2 to 1.5, thereby significantly improving the noise suppression ability.

[0057] In some embodiments, the proportionality coefficient is related to the scale factor of the visual image acquisition device and the cosine value of the angle between the laser plane and the imaging plane; wherein, the scale factor is equal to the product of the pixel size and the object distance divided by the focal length.

[0058] Here, the proportionality coefficient can be the quotient of the scale factor and the cosine value of the angle. For example, the proportionality coefficient can be expressed as s i = where the scale factor of the visual image acquisition device is which is the scale factor when the X-axis of the visual image acquisition device is parallel to the laser plane, and represents the relationship between the object distance, focal length, and pixel size.

[0059] where the cosine value of the angle between the laser plane and the imaging plane is cosθ.

[0060] In this way, the proportionality coefficient can take into account the angle between the laser plane and the imaging plane, avoiding directly using the pixel offset as the true displacement to calculate the proportionality coefficient, which is likely to cause systematic errors.

[0061] In some embodiments, converting the target offset into a physical displacement amount based on the proportionality coefficient includes: multiplying the target offset by the proportionality coefficient to obtain the physical displacement amount.

[0062] This application also provides a large-scale structure deformation laser vision fusion measurement system without calibration, such as Figure 7 shown. The system applies the large-scale structure deformation laser vision fusion measurement method without calibration described in any one or more of the foregoing method embodiments. The system includes: a processing device, and at least a pair of visual image collectors and laser emitters; The at least a pair of visual image collectors and laser emitters are arranged on the inner wall of at least one section of the target structure, and the distance between a pair of visual image collectors and laser emitters is greater than a preset distance; The axis of the laser emitter is parallel to the axis of the target structure, and the processing device is at least communicatively connected to the visual image collector.

[0063] Here, a pair of visual image collectors and laser emitters can be arranged on the inner wall of one section of the target structure, and multiple pairs of visual image collectors and laser emitters can be sequentially arranged on the inner walls of multiple sections of the target structure. The laser emitter uses a semiconductor laser as the light source, combined with precision optical components and an intelligent control circuit. A semiconductor laser with a wavelength of 635nm - 808nm is selected, which can achieve conventional working conditions and high dust penetration, ensuring that the laser beam maintains high collimation during long-distance transmission. The laser emitter is deployed on the center line of the underground structure vault, the embedded base on the side wall or the support member, and is arranged at intervals of 100 - 150m longitudinally along the monitoring area, and forms a network in pairs with the visual image collector (such as an industrial camera, etc.).

[0064] In one embodiment, when installing the laser emitter, it can be fixed by an embedded anchor rod (such as M12 stainless steel) to ensure that the axis of the laser emitter is parallel to the axis of the target structure, and the laser plane can be incident at an angle of 15 - 30° to enhance the fringe contrast. The projection direction extends longitudinally along the structure of the target structure, which is beneficial to covering the full width of the target section (such as from the vault to the invert).

[0065] In one embodiment, the laser emitter may further include: an IP67-level moisture-proof and dust-proof seal, a rubber shock-absorbing pad in the vibration-sensitive area, and a temperature and humidity sensing linkage power adaptive module to ensure the stability of laser projection under complex working conditions.

[0066] In one embodiment, the visual image acquisition device, as the core sensing unit of the system, can be composed of a main camera, an auxiliary camera and a micro weather station to achieve all-round acquisition of deformation information. The main camera can be equipped with a fixed-focus industrial lens with a focal length of 75 mm, combined with a global shutter Complementary Metal-Oxide-Semiconductor (CMOS) sensor, with a resolution of 4096×3072 and a dynamic range of 60 dB. Combined with the high-contrast optical characteristics of the lens, it can clearly capture the deformation details of sub-millimeter laser stripes within a monitoring distance of 20 m.

[0067] In one embodiment, the installation inclination angle of the main camera is fixed at 15°±2°, the baseline distance from the laser emitter is 1.2 m, and a magnetic damping base is used to suppress environmental vibration; the auxiliary camera is arranged at an angle of 30° with the axis of the main camera, and the coaxiality error between the axis of the auxiliary camera and the axis of the main camera is <0.1°. For example, this error can also be calibrated by a theodolite. The field of view overlap rate between the auxiliary camera and the main camera can be greater than or equal to 80%. The overall layout scheme significantly improves the robustness and data reliability of deformation monitoring in complex environments through high-precision optical configuration and strict installation specifications.

[0068] In one embodiment, the processing device can be an intelligent processing module, which undertakes data processing, feature extraction and decision control functions and adopts a three-level architecture design. The processing device can include a field controller, a communication module and an edge server.

[0069] Among them, the field controller can support the real-time operation of deep learning algorithms to complete the sub-pixel positioning of the laser line and the preliminary deformation calculation. The edge server can select an Intel Xeon W-11955M processor, equipped with 128 GB of memory and a 2 TB Solid State Drive (SSD) to achieve parallel processing and long-term storage of multi-section data. The communication module can adopt 5G New Radio (NR) Sub-6GHz technology, supporting a downlink rate of 2 Gbps and an uplink rate of 1 Gbps to ensure the real-time transmission of monitoring data. High-speed data interaction can be achieved between the field controller, the communication module and the edge server through FibreChannel (FC) and Gigabit Ethernet (GbE), with a delay of less than 10 μs. The system adopts containerized deployment technology, supports dynamic expansion and hot update of functional modules, and ensures continuous and stable operation under complex working conditions.

[0070] In some embodiments, the visual image acquisition device includes: a main camera and an auxiliary camera; the angle between the axis of the main camera and the axis of the auxiliary camera is less than a preset threshold.

[0071] Here, the preset threshold value can be 30°, 45°, 60°, etc. For example, the included angle between the main camera axis and the auxiliary camera axis can be 30°.

[0072] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation manner. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc), and includes several instructions for causing a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present application.

[0073] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, a computer, a server, or a data center to another website, a computer, a server, or a data center in a wired manner (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or a wireless manner (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that the computer can access, or a data storage device such as a server or a data center that includes one or more integrated available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium such as a solid state disk (SSD).

[0074] The above are only the preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application are all included in the protection scope of the present application.

[0075] The above description is only a specific implementation manner of the present application, enabling those skilled in the art to understand or implement the present application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but rather will conform to the broadest scope consistent with the principles and novel features claimed herein.

Claims

1. A laser vision fusion measurement method for deformation of large-scale structures without calibration, characterized in that, The method includes: Collecting a target image formed by a laser line emitted by a laser emitter in a target structure through a visual image acquisition device; Determining a target offset of pixel points between the laser line and a reference line based on the target image; Obtaining a proportionality coefficient corresponding to the target offset by querying a predetermined relationship table; the predetermined relationship table records the corresponding relationship between the target offset and the proportionality coefficient; Converting the target offset into a physical displacement based on the proportionality coefficient; Determining deformation data of the target structure according to the physical displacements corresponding to all pixel points.

2. The method according to claim 1, wherein Before collecting a target image formed by a laser line emitted by a laser emitter in a target structure through a visual image acquisition device, the method further includes: Collecting a measurement image formed during a moving process of a laser line emitted by a laser emitter based on a predetermined number of movements of a measurement object through a visual image acquisition device; Performing fitting based on the number of movements of the measurement object, the actual displacement of the measurement object, and the pixel offset in the measurement image to obtain a fitting function between the proportionality coefficient and the pixel offset; Determining the corresponding relationship between the proportionality coefficient and the pixel offset based on the fitting function; Recording the corresponding relationship in a predetermined relationship table.

3. The method according to claim 2, wherein The performing fitting based on the number of movements of the measurement object, the actual displacement of the measurement object, and the pixel offset in the measurement image includes: Performing fitting based on the number of movements of the measurement object, the actual displacement of the measurement object in each movement, the proportionality coefficient corresponding to the change in the angle between the laser plane and the imaging plane caused by each movement, and the pixel offset in the measurement image.

4. The method according to claim 2, wherein There are multiple measurement objects, and the moving direction of the measurement objects is consistent with the deformation direction along the radial direction of the target structure.

5. The method according to claim 2, wherein The determining the corresponding relationship between the proportionality coefficient and the pixel offset based on the fitting function includes: Constructing the Euclidean norm of the residual vector in the fitting function based on the least squares method; Taking the derivative of the Euclidean norm, and determining the corresponding relationship between the proportionality coefficient and the pixel offset based on the result that the derivative is equal to 0.

6. The method according to claim 1, characterized in that The determining the target offset of pixel points between the laser line and the reference line based on the target image includes: After denoising the target image, determining whether the background interference complexity in the target image reaches a predetermined condition; In response to the background interference complexity reaching the predetermined condition, adjusting the threshold corresponding to the target image based on the local window mean, standard deviation, and empirical coefficient corresponding to the target image; Determining the pixel coordinates of the laser line in the target image based on a predetermined algorithm and the threshold; Determining the target offset of pixel points between the laser line and the reference line based on the pixel coordinates.

7. The method according to claim 1, characterized in that, The proportionality coefficient is related to the scale factor of the visual image acquisition device and the cosine value of the angle between the laser plane and the imaging plane; wherein, the scale factor is equal to the product of the pixel size and the object distance divided by the focal length.

8. The method according to claim 1, characterized in that, The converting the target offset into a physical displacement based on the proportionality coefficient includes: Multiplying the target offset by the proportionality coefficient to obtain a physical displacement.

9. A large-scale structure deformation laser vision fusion measurement system without calibration, characterized in that, The system applies the method according to any one of claims 1 to 8, and the system includes: a processing device, and at least a pair of visual image collectors and a laser emitter; The at least a pair of visual image collectors and the laser emitter are arranged on the inner wall of at least one section of the target structure, and the distance between a pair of visual image collectors and the laser emitter is greater than a preset distance; The axis of the laser emitter is parallel to the axis of the target structure, and the processing device is at least communicatively connected to the visual image collector.

10. The system according to claim 9, wherein, The visual image collector includes: a main camera and an auxiliary camera; The angle between the axis of the main camera and the axis of the auxiliary camera is less than a preset threshold.

Images