A method of detecting a deformation of a container

Abstract

This invention provides a method for detecting container deformation, relating to the field of container measurement technology. The method includes: dividing the container into its entire circumference based on its structural characteristics to obtain sub-region division results; performing three-dimensional laser scanning on each sub-region sequentially to obtain a three-dimensional point cloud model of the container; and fitting the projected area and measuring the indentation depth of the deformation region on the three-dimensional point cloud model to obtain the projected area parameters and deformation depth parameters of the deformation region. Addressing the problem of accurately identifying and locating deformation sites, especially small deformations, in containers used in the petrochemical and refining industries, this invention employs a method of first dividing the container into regions and then dynamically stitching them together. This first identifies the deformation location within each sub-region, and then, through dynamic stitching of the sub-regions, identifies the deformation site across the entire container area.

Description

Technical Field

[0001] This invention relates to the field of container measurement technology, and is applied to non-contact measurement and detection methods. Specifically, it relates to a method for detecting container deformation. Background Technology

[0002] Large vertical oil tanks are crucial equipment for oil storage. During their service life, uneven pressure on the tank walls and uneven settlement of the tank foundation can easily lead to varying degrees of geometric deformation. This deformation, in turn, can induce uneven pressure on the tank walls, accelerating tank deformation and potentially causing major safety accidents, resulting in casualties and environmental pollution. Currently, geometric deformation detection has been included as an important part of the annual inspection of oil tanks.

[0003] Currently, conventional methods for detecting tank deformation include: for small-volume metal storage tanks, using a ladder and measuring tape to measure dents and other deformations; however, this method cannot accurately measure the volume of dents and depressions in tank walls with large deformations. Safety inspection of floating roof tanks typically uses optical reference methods and photoelectric methods to measure tank deformation. Optical reference methods require optical plumb bobs, horizontal rulers, and mobile magnetic rulers, but this single-point measurement method can introduce significant measurement errors when the tank wall deformation is complex. Photoelectric methods for measuring tank deformation generally use total stations and GPS, but this method is still a single-point measurement, meaning it can only obtain deformation data from a limited number of observation points to infer the overall deformation. It cannot obtain precise deformation details both locally and globally. Furthermore, the equipment is difficult to install on-site, and the measurement time is relatively long, thus requiring high environmental stability. Its measurement accuracy and stability are often significantly affected, resulting in insufficient detection efficiency and accuracy, failing to meet actual production needs. In addition, existing technologies require measuring numerous geometric parameters and are complex to operate.

[0004] In recent years, the German National Institute of Metrology (PTB) and Trimble Inc. of the United States have developed a novel volume calibration method for the Trimble CX 3D laser scanner, which has been certified by the PTB. This method first preprocesses the point cloud of the tank, then constructs a triangular mesh structure corresponding to the tank. Next, cross-sections of the tank are taken at specified elevations, and the tank volume is calculated using the formula for calculating the volume of a cylinder. However, this method simplifies the tank model, treating it as a regular cylinder, and cannot accurately reflect the geometric deformation information of the tank, resulting in significant errors compared to a realistic tank model.

[0005] To address the problems of existing technologies, this invention provides a method for detecting container deformation. Summary of the Invention

[0006] To address the problems of existing technologies, this invention provides a method for detecting container deformation, the method comprising:

[0007] S1. Based on the structural characteristics of the container under test, the container's entire circumference is divided to obtain the sub-region division results of the container under test;

[0008] S2. Based on the sub-region division results, perform three-dimensional laser scanning on each sub-region in sequence to obtain a three-dimensional point cloud model of the container to be tested;

[0009] S3. For the three-dimensional point cloud model, the projected area of ​​the deformation region is fitted and the depression depth is measured to obtain the projected area parameters and deformation depth parameters of the deformation region.

[0010] According to an embodiment of the present invention, step S1 includes:

[0011] Obtain the original structural data of the container under test to obtain the structural features of the container under test;

[0012] Let the diameter of the container to be measured be D. Divide the container into M equal parts along the circumference, satisfying the following condition: Where m is the circumferential direction division coefficient;

[0013] Let the height of the container to be measured be H. Divide the container into N equal parts along the height direction, satisfying the following condition: Where n is the height direction division coefficient;

[0014] After dividing the area in the circumferential direction and the height direction, the surface of the container under test is divided into Q sub-regions, where Q = M × N;

[0015] Sub-regions at the same horizontal level are numbered in a clockwise order along the circumference, and sub-regions at the same vertical line are numbered in a bottom-up order along the height, resulting in the sub-region division of the container under test. Each sub-region on the circumferential surface of the container is numbered (M...). i N k ), i = 1 to M, k = 1 to N.

[0016] According to an embodiment of the present invention, step S1 further includes setting splicing identification points at the boundaries of each sub-region based on the sub-region division result, which includes:

[0017] Along the circumference of each sub-region, a splicing identification point is set at intervals of a first distance, wherein the first distance is... q is the coefficient of the marker point;

[0018] Along the height direction of each sub-region, a stitching recognition point is set at every second distance, where the second distance...

[0019] When setting up splicing recognition points, only one set of splicing recognition points needs to be set at the boundary of every two adjacent sub-regions.

[0020] According to an embodiment of the present invention, step S2 includes:

[0021] Based on the sub-region division results, three-dimensional laser scanning is performed on each sub-region in sequence to obtain the sub-region point cloud data set of the container under test;

[0022] Based on the sub-region point cloud data set and combined with the stitching recognition points, the container under test is dynamically stitched around its entire circumference to obtain the three-dimensional point cloud model of the container under test.

[0023] According to an embodiment of the present invention, the sub-region point cloud data set of the container under test is obtained through the following steps:

[0024] a. For the sub-region to be scanned, set up the scanning device and scan the sub-region in a scanning order from bottom to top and from left to right;

[0025] b. Adjust the relative angle between the scanning field of view and the surface of the sub-region to be scanned, and return to step a after each adjustment;

[0026] c. After the scan is completed, the point cloud data set of the sub-region of the container to be tested is obtained.

[0027] According to one embodiment of the present invention, full-circle dynamic stitching is performed through the following steps:

[0028] Based on the sub-region point cloud data set, the point cloud data of each sub-region is cropped, and the point cloud data outside each sub-region is reduced to obtain the cropping result data of each sub-region.

[0029] For each sub-region, the cropping result data is optimized using point cloud data to obtain the optimized result data for each sub-region;

[0030] Based on the optimization results data and combined with the stitching recognition points, the point cloud data between sub-regions is dynamically stitched together to obtain the three-dimensional point cloud model of the entire circumference of the container under test.

[0031] According to an embodiment of the present invention, the cropping result data of each sub-region is obtained through the following steps:

[0032] Let Q be the set of point cloud data corresponding to the current sub-region. all ;

[0033] Using the stitched recognition points corresponding to the current sub-region as the boundary, mark the point cloud data set outside the boundary as Q. out The point cloud data set within the boundary is Q. in;

[0034] Perform point cloud dataset computation Q on the current sub-region out =Q all -Q in The point cloud data outside the current sub-region is clipped to obtain the clipping result data corresponding to the current sub-region.

[0035] According to an embodiment of the present invention, the optimized result data for each sub-region is obtained through the following steps:

[0036] Let O be the central axis of the container to be measured. Then the deformation of the container to be measured perpendicular to the central axis O is Δd, where Δd is the height of the protrusion or the depth of the depression of the container to be measured.

[0037] For the case of convex deformation, the point cloud at the farthest vertical distance from the central axis O is denoted as [the distance from the central axis O is denoted as ].

[0038] Let R1 be the number of point clouds at the convex tip. If R1 satisfies the formula: S is the threshold for the number of point clouds, then it is determined that... Within the scope, with Data within the radius is considered to be randomly occurring point cloud data and is deleted; K is the radius coefficient.

[0039] For concave deformation, the point cloud at the nearest vertical distance from the central axis O has a distance of [missing information].

[0040] Let R2 be the number of point clouds at the apex of the depression. If R2 satisfies the formula: Then it is determined Within the scope, with Data within the radius is considered to be accidental point cloud data and is deleted.

[0041] According to an embodiment of the present invention, the three-dimensional point cloud model of the entire circumference of the container under test is obtained by stitching together the following steps:

[0042] Sub-region (M) i N k ) and sub-region (M) i+1 N k The splicing error at each circumferential splicing recognition point between ) is Δe i Then the cumulative error between the two regions in the circumferential direction is

[0043] Sub-region (M) i N k ) and sub-region (M) i N k+1The stitching error at each height stitching recognition point between the two points is Δe. k Then the cumulative error between the two regions in the height direction is

[0044] Compute sub-region (M) i N k The cumulative error of the splicing recognition points is like δ is the stitching error threshold, then the sub-region (M) is considered to be... i N k Successfully joined with adjacent sub-regions;

[0045] Subregion (M) i N k After successful stitching, move to the next stitching sub-region (M). i+1 N k ) or (M i N k+1 ), determine sub-region (M) i+1 N k ) or (M i N k+1 Whether it is successfully stitched with the adjacent sub-regions, and so on, until all sub-regions are successfully stitched together, to obtain the three-dimensional point cloud model.

[0046] According to an embodiment of the present invention, in step S3, the projected area of ​​the deformation region is fitted through the following steps:

[0047] When the three-dimensional point cloud model is unfolded circumferentially, the longer side of the unfolded rectangle is the perimeter of the container, marked as the X-axis, and the shorter side of the unfolded rectangle is the height of the container, marked as the Y-axis.

[0048] Calculate the average value of all point cloud data in the unfolded rectangle along the Z-axis, and mark it as the base plane of the unfolded rectangle;

[0049] The portions above and below the base plane are respectively denoted as convex deformation and concave deformation. The process begins by traversing the point cloud data farthest from the base plane, continuing until the Z-coordinate value of the traversed point cloud data satisfies: Z... i -Z O If Z ≤ ∈, then stop traversing, where Z i Z is the Z-coordinate value of the i-th point cloud data. O is the Z-coordinate value of the point cloud data at the base plane, and ∈ is the deformation threshold;

[0050] Record the projected contour enclosed by the point cloud where the traversal stops, then the projected contour is the range where the deformation occurs;

[0051] The projected contour is fitted using geometric shapes, and the area of ​​the deformed contour image obtained after fitting is the projected size of the deformed region.

[0052] According to an embodiment of the present invention, step S3 further includes: identifying the location of the deformation occurrence region based on the projected area parameter of the deformation region, which includes:

[0053] The shape contour of the deformed region is determined by the projected area parameter, and the centroid position of the shape contour of the deformed region is marked.

[0054] The sub-region where the centroid is located is marked as the location where the deformation occurs.

[0055] According to an embodiment of the present invention, in step S3, the depth of the depression in the deformation region is measured through the following steps:

[0056] Select the deformation contour range of the concave area in the deformation contour image;

[0057] Determine the point cloud that is farthest from the base plane within the deformation profile range of the depression, and read the distance Z of this point cloud from the base plane in the Z-axis direction. imax Record Z imax The depth of the depression in the deformation area.

[0058] According to another aspect of the invention, a storage medium is also provided, which includes a series of instructions for performing the steps of the method described in any of the preceding claims.

[0059] According to another aspect of the invention, an apparatus for detecting container deformation is also provided, performing the method as described in any of the preceding claims, the apparatus comprising:

[0060] The sub-region division module is used to divide the container around the test container according to the structural characteristics of the container, and obtain the sub-region division result of the container.

[0061] The three-dimensional point cloud model module is used to perform three-dimensional laser scanning on each sub-region in sequence based on the sub-region division results to obtain the three-dimensional point cloud model of the container to be tested.

[0062] The module for projected area parameters and deformation depth parameters is used to fit the projected area of ​​the deformation region and measure the depression depth for the three-dimensional point cloud model, so as to obtain the projected area parameters and deformation depth parameters of the deformation region.

[0063] This invention provides a method for detecting container deformation, which has the following advantages compared with the prior art:

[0064] 1) To address the problem of accurately identifying and locating deformed parts, especially small deformed parts, of containers in the petrochemical and refining fields, this invention uses a method of first dividing the area and then dynamically splicing it together. This method first identifies the deformation location within each sub-region, and then, through dynamic splicing of the sub-regions, identifies the deformed part on the entire container area.

[0065] 2) To address the difficulty in accurately quantifying the deformation of containers in the petrochemical and refining industries, this invention proposes a three-dimensional measurement method using laser structured light for measuring deformation on container surfaces. This method quantifies the degree of deformation from two perspectives: the projected area parameter of the deformation region and the deformation depth parameter.

[0066] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims, and drawings. Attached Figure Description

[0067] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0068] Figure 1 A flowchart illustrating the steps of a method for detecting container deformation according to an embodiment of the present invention is shown.

[0069] Figure 2 A flowchart illustrating the steps for obtaining the sub-region division result of the container under test according to an embodiment of the present invention is shown.

[0070] Figure 3 A schematic diagram of sub-region division according to an embodiment of the present invention is shown;

[0071] Figure 4 A flowchart illustrating the steps for obtaining a three-dimensional point cloud model of a container under test according to an embodiment of the present invention is shown.

[0072] Figure 5 A flowchart illustrating the steps of full-circle dynamic stitching according to an embodiment of the present invention is shown;

[0073] Figure 6 A flowchart illustrating the steps of a point cloud data optimization method according to an embodiment of the present invention is shown.

[0074] Figure 7 A flowchart illustrating the steps for obtaining the projected area parameters of the deformed region, the location of the deformation, and the deformation depth parameters according to an embodiment of the present invention is shown.

[0075] Figure 8A flowchart illustrating the steps of fitting the projected area of ​​the deformation-occurring region according to an embodiment of the present invention is shown.

[0076] In the accompanying drawings, the same parts use the same reference numerals. Also, the drawings are not drawn to scale. Detailed Implementation

[0077] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0078] Existing technology (CN111504214B) provides a fiber optic monitoring method for deformation of a large crude oil storage tank. The fiber optic monitoring device for large crude oil storage tank deformation includes distributed fiber optic strain sensors and fiber optic grating array strain sensors. The distributed fiber optic strain sensors are spirally arranged on the surface of the tank to be monitored, while the fiber optic grating array strain sensors are fixed to the surface at the location of deformation. When the distributed fiber optic strain sensors are spirally arranged on the tank, the spiral spacing is H, and the spiral angle is θ. The spiral spacing H is 0.3 to 0.4 times the height of each wall panel on the tank, where θ = arctan(H / D), D is the diameter of the tank, and the spiral angle is the angle between the distributed fiber optic strain sensors and the diameter direction of the tank's foundation plane. This method achieves overall deformation monitoring of the tank using distributed fiber optic strain sensors, while simultaneously enabling precise monitoring of local deformation using fiber optic grating array strain sensors.

[0079] However, the existing technology uses the fiber optic method to measure the deformation of large crude oil storage tanks. The method involves setting distributed fiber optic strain sensors in a spiral arrangement on the surface of the tank to be monitored, and fixing fiber optic grating array strain sensors on the surface of the tank where deformation occurs. However, this method requires manpower and time to fix the sensors on the surface of large tanks.

[0080] The existing technology (CN114739311B) is applicable to rapid deformation monitoring of mine shafts. The equipment employs multiple sensors to measure the initial coordinates of a 2D lidar, which is then gradually lowered into the shaft. Data is recorded at different times. The data recorded by the wheeled odometer and inertial measurement unit (IMU) are preprocessed to calculate the lowering distance of the steel wire and the pose data of the IMU at different times. The lowering distance data calculated by the wheeled odometer and the pose data calculated by the IMU are interpolated by timestamps to obtain the pose of the 2D lidar at each timestamp. Precise coordinates of the shaft wall at different depths are calculated, a three-dimensional mesh model of the shaft is established, and deformation data of the shaft is obtained. This method is safe, highly accurate, low-cost, requires less labor, and generates a large amount of measurement data. It reduces and avoids injuries to measurement personnel caused by complex environments during shaft measurement, greatly improving the efficiency of shaft monitoring.

[0081] However, this existing technology obtains deformation data indirectly and cannot directly determine the location and extent of deformation.

[0082] The prior art (CN114074724B) provides a wall-climbing robot applicable to the inner and outer walls of storage tanks with different radii of curvature. It includes a vehicle body unit, directional wheel drive units and omnidirectional wheel follower units symmetrically arranged from front to back on both sides of the vehicle body unit, and a replaceable magnetic wheel unit. The vehicle body unit has a rectangular box-shaped structure. An array of permanent magnets is provided on the lower cover plate of the vehicle body unit, matching the replaceable magnetic wheel unit. A camera mounting box for fixing a front camera is provided on the front side plate of the vehicle body unit, and a rear camera is fixed to the rear of the upper cover plate of the vehicle body unit via a camera mounting plate. This invention, by combining the vehicle body unit, directional wheel drive unit, and omnidirectional wheel follower unit according to the specific connection and positional relationships defined in this invention, can adjust the tilt angle according to different wall surface conditions, ensuring the effective adsorption area and distance of the permanent magnets on the tank surface, thus improving the robot's adsorption effectiveness and safety on surfaces with different curvatures.

[0083] However, this existing technology has limitations in dealing with steeply shaped depressions on the facade surface, where the robot cannot be fixed in place and therefore cannot cross the depression defects.

[0084] Existing technology (CN111854623B) provides a rapid detection method and system for detecting minute deformations of objects. Existing technology (CN114114107B) provides an experimental device for measuring magnetostrictive minute deformations. Existing technology (CN114878320B) provides a deformation detection mechanism, a pressure vessel detection device, and a detection system. Existing technology (CN103335588B) provides... Existing technology (Research on Measurement of Deformation of Loaded Pressure Vessel Wall Using Digital Speckle Method. Automation and Instrumentation 2012(2),2.) provides research on measuring the deformation of loaded pressure vessel wall using digital speckle method. Existing technology (Laser Overlapping 3D Reconstruction of Damaged Aero-engine Blades. Progress in Laser and Optoelectronics, 2020, 57(16),9.) provides laser overlapping 3D reconstruction of damaged aero-engine blades. However, the above-mentioned existing technologies cannot simultaneously achieve the problem of measurement function and spatial resolution for detecting the deformation of large, medium, and small containers.

[0085] In summary, existing non-contact detection technologies face the challenge of simultaneously possessing both measurement capabilities and spatial resolution for detecting the deformation of large, medium, and small containers.

[0086] To address the aforementioned shortcomings of existing technologies and the difficulty in accurately identifying and locating deformed parts, especially small deformities, in tanks used in the petrochemical and refining industries, this invention employs a method of first dividing the tank into regions and then dynamically splicing them together. This method first identifies the deformation location within each sub-region and then, through dynamic splicing of the sub-regions, identifies the deformed parts across the entire tank area.

[0087] To address the aforementioned shortcomings of existing technologies and the difficulty in accurately quantifying the deformation of tank bodies in the petrochemical and refining industries, this invention proposes a three-dimensional measurement method using laser structured light for measuring surface deformation of tank bodies. This method quantifies the degree of deformation from two perspectives: the projected area parameter of the deformation region and the deformation depth parameter.

[0088] Therefore, this invention solves the problem of traditional tank deformation measurement technology, which can only obtain deformation data from a limited number of observation points and thus infer the overall deformation, enabling accurate acquisition of precise local and overall deformation details. This invention also addresses the technical problems of low efficiency and low measurement accuracy in tank inspection. Furthermore, this invention solves the problem of indirectly acquiring deformation data by directly identifying the location of tank deformation and quantitatively describing the degree of deformation.

[0089] Figure 1 A flowchart illustrating the steps of a method for detecting container deformation according to an embodiment of the present invention is shown.

[0090] like Figure 1 As shown, in step S1, based on the structural characteristics of the container under test, the container's entire circumference is divided to obtain the sub-region division results. Specifically, in step S1, for the container under test, through methods such as... Figure 2 The container's full-circle region division method shown yields the sub-region division results of the container under test.

[0091] Figure 2 A flowchart illustrating the steps for obtaining the sub-region division result of the container under test according to an embodiment of the present invention is shown.

[0092] like Figure 2 As shown, in step S201, the original structural data of the container under test is obtained. Specifically, the original structural data of the container under test is obtained to obtain the structural features of the container under test. Further, the original structural data ledger of the container under test is obtained, wherein the original structural data ledger of the container specifically includes: container diameter parameters and container height parameters.

[0093] like Figure 2 As shown, in step S202, the circumferential direction is divided. Specifically, let the diameter of the container to be measured be D, and divide the container to be measured into M equal parts along the circumference, satisfying... Where m is the circumferential division coefficient. Further, m = 1.8, m = 2.4, m = 3, or m = 3.6, with units of length in SI units.

[0094] like Figure 2 As shown, in step S203, the height direction is divided. Specifically, let the height of the container to be measured be H, and divide the container to be measured into N equal parts along the height direction, satisfying... Where n is the height division coefficient. Further, n = 1.8, n = 2.4, n = 3, or n = 3.6, and the unit is the International System of Units (SI) for length.

[0095] like Figure 2 As shown, in step S204, the sub-regions are sequentially labeled. Specifically, after dividing the surface of the container under test into circumferential and vertical regions, the cylindrical surface is divided into Q sub-regions, where Q = M × N. Further, sub-regions at the same horizontal level are numbered in a clockwise direction in the circumferential direction, and sub-regions at the same vertical line are numbered in a bottom-up direction in the vertical direction, resulting in the sub-region division of the container under test. Each sub-region on the circumferential surface of the container under test is numbered (M... i N k ), i = 1 to M, k = 1 to N.

[0096] Figure 3 A schematic diagram of sub-region division according to an embodiment of the present invention is shown.

[0097] like Figure 3 As shown, in step S204, the first sub-region of interest is denoted as (M1, N1), and starting from the first sub-region, the regions to be measured at the same horizontal level are denoted as (M1, N1) clockwise. i (M1, N1), where i = 1 to M; taking the first sub-region as the starting point, sub-regions at the same vertical height from bottom to top are denoted as (M1, N2, N3, N4, N5, N6, N7, N8, N9, N1, N9, N1, N2 ... k ), where k = 1 to N. Therefore, each sub-region of the circumferential surface of the measured container is denoted as (M i N k ), i = 1 to M, k = 1 to N.

[0098] In one embodiment, step S1 further includes setting splicing identification points at the boundaries of each sub-region based on the sub-region division results. Specifically, splicing identification points are set at the boundaries of each sub-region through the following steps: along the circumferential direction of each sub-region, a splicing identification point is set at intervals of a first distance, wherein the first distance is... q represents the marker coefficient; along the height direction of each sub-region, a stitching identification point is set at every second distance, where the second distance... When setting the splicing identification points, only one set of splicing identification points needs to be set at the boundary of every two adjacent sub-regions. Furthermore, q = 3, q ​​= 4, q = 5, or q = 6.

[0099] like Figure 1 As shown, in step S2, based on the sub-region division results, each sub-region is sequentially subjected to 3D laser scanning to obtain a 3D point cloud model of the container under test. Specifically, in step S2, based on the sub-region division results, through methods such as... Figure 4 The method for locating the localized deformed parts and the method for dynamic splicing of the container around its entire circumference, as shown, are used to obtain a three-dimensional point cloud model of the container under test.

[0100] Figure 4 A flowchart illustrating the steps for obtaining a three-dimensional point cloud model of a container under test according to an embodiment of the present invention is shown.

[0101] like Figure 4 As shown, in step S401, based on the sub-region division results, three-dimensional laser scanning is performed on each sub-region sequentially to obtain a set of sub-region point cloud data for the container under test. Specifically, step S401 is a method for locating locally deformed parts, and the set of sub-region point cloud data for the container under test is obtained through the following steps:

[0102] a. For the sub-region to be scanned, set up the scanning device and scan the sub-region in a scanning order from bottom to top and from left to right. Specifically, for the selected sub-region (M) i N k The striped structured light is set at a certain distance from the surface of the area to be measured, and the sub-regions are scanned in a scanning order from top to bottom (or from bottom to top) and from left to right (or from right to left).

[0103] b. Adjust the relative angle between the scanning field of view and the surface of the sub-region to be scanned, and return to step a after each adjustment. Specifically, adjust the relative angle between the scanning field of view and the surface of the region to be measured, and execute steps a-b after each adjustment.

[0104] c. After scanning, the point cloud data set of the sub-region of the container under test is obtained. Specifically, the sub-region under test (M) is obtained. i N k The complete point cloud data set within the container to be tested is used to obtain the sub-region point cloud data set of the container to be tested.

[0105] like Figure 4 As shown, in step S402, based on the sub-region point cloud data set and combined with the stitching recognition points, the container under test is dynamically stitched around its entire perimeter to obtain a 3D point cloud model of the container. Specifically, step S402 is a method for dynamic stitching of the container around its entire perimeter. Further, through methods such as... Figure 5 The method shown performs full-circle dynamic stitching.

[0106] Figure 5 A flowchart illustrating the steps of full-circle dynamic stitching according to an embodiment of the present invention is shown.

[0107] like Figure 5 As shown, in step S501, based on the sub-region point cloud data set, the point cloud data of each sub-region is clipped to reduce the point cloud data outside each sub-region, thereby obtaining the clipping result data of each sub-region.

[0108] Specifically, the cropping results of each sub-region are obtained through the following steps: Record the current sub-region (M) i N k The corresponding point cloud dataset is Q. all Using the stitching recognition points corresponding to the current sub-region as the boundary, mark the point cloud data set outside the boundary as Q. out The point cloud data set within the boundary is Q. in ; Perform point cloud dataset calculation Q for the current sub-region out =Q all -Q in The point cloud data outside the current sub-region is clipped to obtain the clipping result data corresponding to the current sub-region.

[0109] like Figure 5 As shown, in step S502, point cloud data optimization is performed on the cropping result data for each sub-region to obtain the optimized result data for each sub-region. Specifically, in step S502, for each sub-region, point cloud data optimization is performed using methods such as... Figure 6 The point cloud data optimization method shown yields the optimization results for each sub-region.

[0110] Figure 6 A flowchart illustrating the steps of a point cloud data optimization method according to an embodiment of the present invention is shown.

[0111] like Figure 6 As shown, the optimized result data for each sub-region is obtained through the following steps: Let the central axis of the container under test be O, then the deformation of the container perpendicular to the central axis O is Δd, where Δd is the convex height or concave depth of the container under test; for convex deformation, the point cloud at the farthest vertical distance from the central axis O is denoted as [missing information - likely a typo]. Let R1 be the number of point clouds at the convex tip. If R1 satisfies the formula: S is the threshold for the number of point clouds, then it is determined that... Within the scope, with Data within the radius is considered accidental point cloud data and is deleted; K is the radius coefficient. For concave deformation, the point cloud at the nearest vertical distance from the central axis O has a distance of [missing information]. Let R2 be the number of point clouds at the apex of the depression. If R2 satisfies the formula: Then it is determined Within the scope, with Data within the radius is considered accidental point cloud data and is deleted. Further, S = 1-100, where S is an integer. K = 5, K = 6, K = 7, K = 8, K = 9, or K = 10.

[0112] like Figure 5 As shown, in step S503, based on the optimization result data and combined with the splicing recognition points, the point cloud data between sub-regions is dynamically spliced ​​to obtain a three-dimensional point cloud model of the entire circumference of the container under test.

[0113] Specifically, the following steps are used to stitch together a 3D point cloud model of the entire circumference of the container under test: (M) i N k ) and sub-region (M) i+1 N k The splicing error at each circumferential splicing recognition point between ) is Δe i Then the cumulative error between the two regions in the circumferential direction is Sub-region (M) i N k ) and sub-region (M) i N k+1 The stitching error at each height stitching recognition point between the two points is Δe. k Then the cumulative error between the two regions in the height direction is Compute sub-region (M) i N k The cumulative error of the splicing recognition points is like δ is the stitching error threshold, then the sub-region (M) is considered to be... i N k The sub-region (M) was successfully joined with the adjacent sub-region; i N k After successful stitching, move to the next stitching sub-region (M). i+1 N k ) or (M i N k+1 ), determine sub-region (M) i+1 N k ) or (M i N k+1 Whether the stitching with adjacent sub-regions is successful is checked, and so on, until all sub-regions are successfully stitched together to obtain a 3D point cloud model. Further, δ = 1 / 20r, where r is the radius of the stitching recognition point.

[0114] like Figure 1As shown, in step S3, for the 3D point cloud model, the projected area of ​​the deformation region is fitted and the depression depth is measured to obtain the projected area parameters and deformation depth parameters of the deformation region. Specifically, step S3 also includes: identifying the location of the deformation region based on the projected area parameters of the deformation region. Further, in step S3, for the 3D point cloud model, through methods such as... Figure 7 The method shown yields the projected area parameters of the deformed region, the location of the deformation, and the deformation depth parameters.

[0115] Figure 7 A flowchart illustrating the steps for obtaining the projected area parameters of the deformed region, the location of the deformation, and the deformation depth parameters according to an embodiment of the present invention is shown.

[0116] like Figure 7 As shown, in step S701, the projected area of ​​the deformation region is fitted. Specifically, step S701 is a method for fitting the projected area of ​​the deformation region, further, through methods such as... Figure 8 The method shown is used to fit the projected area of ​​the deformation region.

[0117] Figure 8 A flowchart illustrating the steps of fitting the projected area of ​​the deformation-occurring region according to an embodiment of the present invention is shown.

[0118] like Figure 8 As shown, in step S801, the 3D point cloud model is unfolded circumferentially. Specifically, when the 3D point cloud model is unfolded circumferentially, the longer side of the unfolded rectangle is the container perimeter, marked as the X-axis, and the shorter side of the unfolded rectangle is the container height, marked as the Y-axis. The name of the unfolded sub-region (M) is... i N k No changes will be made.

[0119] like Figure 8 As shown, in step S802, a base plane in the Z direction is constructed. Specifically, the average value of all point cloud data in the unfolded rectangle along the Z axis is calculated and marked as the base plane of the unfolded rectangle. That is, the average value of the Z-direction coordinates of all point cloud data in the rectangle is taken and denoted as the base plane in the Z direction of the unfolded rectangle.

[0120] like Figure 8 As shown, in step S803, the point cloud data is traversed along the Z-axis base plane. Specifically, the portions above and below the base plane are denoted as convex deformation and concave deformation, respectively. The traversal begins with the point cloud data furthest from the base plane and continues until the Z-coordinate values ​​of the traversed point cloud data satisfy: Z i -Z O If ≤∈(step S804), then stop the traversal (step S805), where Z iZ is the Z-coordinate value of the i-th point cloud data. O Let Z be the Z-coordinate value of the point cloud data at the base plane, and ∈ be the deformation threshold. Further, ∈ = 0.01 or ∈ = 0.1 or ∈ = 0.2 or ∈ = 0.5.

[0121] like Figure 8 As shown, in step S806, the projected contour enclosed by the point cloud where traversal stops is recorded. Specifically, recording the projected contour enclosed by the point cloud where traversal stops means that the projected contour represents the range where deformation occurs.

[0122] like Figure 8 As shown, in step S807, a geometric shape is used to fit the projected contour. Specifically, the area of ​​the image obtained after fitting the geometric shape is the projected size of the deformation region. Further, the geometric shape used to fit the projected contour can be any one or any combination of rectangles, circles, and triangles.

[0123] like Figure 7 As shown, in step S702, the deformation location is identified. Specifically, step S702 is a method for identifying the location of the deformation region, which involves the following steps: determining the shape contour of the deformation region using the projected area parameter, marking the centroid position of the shape contour of the deformation region; and marking the sub-region where the centroid position is located as the deformation location.

[0124] like Figure 7 As shown, in step S703, the depth of the depression in the deformation area is measured. Specifically, step S703 is a method for measuring the depth of the depression in the deformation area, which involves the following steps: Figure 3 As shown, the deformation contour range of the depression in the deformation contour image is selected; the point cloud farthest from the base plane within the deformation contour range of the depression is determined, and the distance Z of this point cloud from the base plane in the Z-axis direction is read. imax Record Z imax The depth of the depression in the deformation region. Further, based on the unfolded rectangular 3D point cloud surface, the deformation contour range at the depression is selected. Within the range enclosed by the deformation contour, the point cloud element furthest from the base plane is selected, and its Z-direction distance Z from the base plane is read. imax This refers to the depth at which the concave deformation occurs.

[0125] This invention employs a non-contact optical detection method. During detection, the device is positioned at a certain distance from the tank being tested and its surface is scanned and imaged using a three-dimensional laser scanning method. This method is unaffected by the material of the tank or any surface protrusions or depressions.

[0126] This invention employs a two-step method combining location identification with three-dimensional dimension measurement of the deformed area to achieve precise positioning of the deformed area; and quantitatively determines the degree of deformation based on the depth of the depression and the projected area of ​​the deformed edge.

[0127] This invention employs a non-contact 3D detection method, eliminating the need for wiring and sensor installation. By dividing the tank into multiple sub-regions and setting splicing recognition points, dynamic splicing is performed based on these points to combine the sub-regions into a complete 3D model of the tank, thereby enabling the measurement of the 3D tank diameter and local deformation.

[0128] This invention can be widely applied to the optical detection and measurement of deformation of large and medium-sized equipment in the petroleum and chemical industries. The method is widely applicable to the materials of the equipment itself. The proposed detection method based on the measurement of three-dimensional deformation of equipment is conducive to quickly locating the deformed parts and quantitatively evaluating the three-dimensional geometric features of the deformed parts, which is of great significance for the evaluation technology of equipment safe service.

[0129] The method for detecting container shape variables provided by this invention can also be used in conjunction with a computer-readable storage medium storing a computer program. Executing the computer program runs the method for detecting container shape variables. The computer program can execute computer instructions, which include computer program code. The computer program code can be in the form of source code, object code, executable files, or some intermediate form.

[0130] Computer-readable storage media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0131] It should be noted that the contents of computer-readable storage media may be appropriately added to or subtracted from the contents according to the requirements of legislation and patent practice in a jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable storage media may not include electrical carrier signals and telecommunication signals.

[0132] According to another aspect of the present invention, an apparatus for detecting container deformation is also provided, which performs a method for detecting container deformation. The apparatus includes: a sub-region division module, a three-dimensional point cloud model module, and a projection area parameter and deformation depth parameter module.

[0133] The sub-region segmentation module is used to divide the entire circumference of the container under test according to its structural features, obtaining the sub-region segmentation results. The 3D point cloud model module is used to perform 3D laser scanning on each sub-region based on the sub-region segmentation results, obtaining the 3D point cloud model of the container under test. The projected area parameter and deformation depth parameter modules are used to fit the projected area of ​​the deformation area and measure the indentation depth for the 3D point cloud model, obtaining the projected area parameters and deformation depth parameters of the deformation area.

[0134] In one embodiment, a gas buffer tank with a diameter of 7m and a height of 7m is used as the container to be tested. The method for detecting container deformation provided by this invention is used to detect the degree of deformation of its cylindrical surface around its entire circumference. The steps are as follows:

[0135] S1. Divide the gas buffer tank into 12 equal parts along the circumference;

[0136] S2. Divide the gas buffer tank into 4 equal parts along the height direction;

[0137] S3. After division, the surface of the gas buffer tank to be tested is divided into 48 sub-regions.

[0138] S4. In each sub-region, a marker point is placed at a distance of 0.3m along the circumferential direction for splicing identification points;

[0139] S5. A marker point is placed at 0.3m intervals for the splicing identification points in the axial direction of each sub-region;

[0140] S6. Using striped structured light at a certain distance from the surface of the gas buffer tank, scan sequentially from sub-region (M1, N1) to sub-region (M... 12 (N4);

[0141] S7. Obtain the point cloud data set of the entire circumference surface of the gas buffer tank cylinder;

[0142] S8. Sequentially divide the sub-regions (M1, N1) to (M... 12 The point cloud data of N4) is cropped to reduce the point cloud data of sub-regions outside the X and Y directions;

[0143] S9. Let Δd be the deformation of the sub-region at a distance perpendicular to the central axis O. Optimize the farthest point cloud data in the Z direction for each sub-region:

[0144] S9.1. Take K = 5 and S = 10;

[0145] S9.2, if

[0146] S9.3, Remove point cloud data within the radius of Δd / 5;

[0147] S9.4 The remaining point cloud data is the optimized point cloud data set;

[0148] S10. The cropped and optimized point cloud data of the 48 sub-regions are stitched together, with a cumulative stitching error of 1.0mm.

[0149] S11. Unfold the 3D container model along the circumference to form a rectangle;

[0150] S12. Take the average value of the Z-direction coordinates of all point cloud data in the rectangle, and denote it as the base plane of the unfolded rectangle;

[0151] S13. Denote the portions above and below the base plane as convex deformation and concave deformation, respectively. Start traversing from the point cloud data furthest from the base plane until the Z-coordinate of the traversed point cloud data satisfies: Z i -Z O If the value is ≤0.5, the iteration stops.

[0152] S14. Record the shape of the projected outline enclosed by the point cloud where the traversal stops. This shape is the range in which the deformation occurs.

[0153] S15. Multiple rectangles are used to fit the projected contour shape, and the area of ​​the image obtained after fitting is the projected size where the deformation occurs.

[0154] S16. Based on the shape contour of the deformed region, mark its centroid position;

[0155] S17. The sub-region where the center of mass is located is the location where deformation occurs.

[0156] S18. Based on the unfolded rectangular 3D point cloud surface, select the deformation contour range of the concave area;

[0157] S19. Select the point cloud that is farthest from the base plane within the area enclosed by the deformation profile, and read its Z-direction distance value from the base plane. imax This refers to the depth at which the concave deformation occurs.

[0158] In one embodiment, a horizontal salt mixer with a diameter of 0.6m and a length of 3.74m is used as the container to be tested. The method for detecting container deformation provided by this invention is used to detect the degree of deformation of its cylindrical surface around its entire circumference. The steps are as follows:

[0159] S1. Divide the horizontal salt mixer into 3 equal parts along the circumference;

[0160] S2. Divide the horizontal salt mixer into 4 equal parts along the height direction;

[0161] S3. After division, the surface of the horizontal salt mixer under test is divided into 12 sub-regions.

[0162] S4. In each sub-region, a marker point is placed at a distance of 0.18m along the circumferential direction for splicing identification points;

[0163] S5. A marker point is placed at 0.18m intervals for the splicing identification points in the axial direction of each sub-region;

[0164] S6. Using striped structured light at a certain distance from the surface of the gas buffer tank, scan sequentially from sub-region (M1, N1) to sub-region (M3, N4).

[0165] S7. Obtain the point cloud data set of the entire circumference surface of the horizontal salt mixer cylinder;

[0166] S8. The point cloud data of sub-regions (M1, N1) to (M3, N4) are clipped sequentially, and the point cloud data of the sub-regions outside the X and Y directions are clipped.

[0167] S9. Let Δd be the deformation of the sub-region at a distance perpendicular to the central axis O. Optimize the farthest point cloud data in the Z direction for each sub-region:

[0168] S9.1. Take K = 10, S = 1;

[0169] S9.2, if

[0170] S9.3, Remove point cloud data within the radius range of Δd / 10;

[0171] S9.4 The remaining point cloud data is the optimized point cloud data set;

[0172] S10. The cropped and optimized point cloud data of the 12 sub-regions are stitched together, with a cumulative stitching error of 0.5mm.

[0173] S11. Unfold the three-dimensional horizontal salt mixer model along the circumference to form a rectangle;

[0174] S12. Take the average value of the Z-direction coordinates of all point cloud data in the rectangle, and denote it as the base plane of the unfolded rectangle;

[0175] S13. Denote the portions above and below the base plane as convex deformation and concave deformation, respectively. Start traversing from the point cloud data furthest from the base plane until the Z-coordinate of the traversed point cloud data satisfies: Z i -Z O If the value is less than or equal to 0.01, the iteration stops.

[0176] S14. Record the shape of the projected outline enclosed by the point cloud where the traversal stops. This shape is the range in which the deformation occurs.

[0177] S15. Unlike the gas buffer tank embodiment, in order to reduce the error in the projected area size, this embodiment uses two geometric shapes, a rectangle and a circle, to fit the projected outline shape. The area of ​​the image obtained after fitting is the projected size where the deformation occurs.

[0178] S16. Based on the shape profile of the deformed region, calculate and mark its centroid position;

[0179] S17. The sub-region where the center of mass is located is the location where deformation occurs.

[0180] S18. Based on the unfolded rectangular 3D point cloud surface, select the deformation contour range of the concave area;

[0181] S19. Select the point cloud that is farthest from the base plane within the area enclosed by the deformation profile, and read its Z-direction distance value from the base plane. imax This refers to the depth of the horizontal container that undergoes concave deformation.

[0182] In one embodiment, a small container (such as an oxygen, nitrogen, or helium cylinder) is used as the container to be tested. The method for detecting container deformation provided by this invention is used to detect the degree of deformation after compression. The steps are as follows:

[0183] First, the gas cylinder to be tested is divided according to its structural regions: the part above the cylinder hoop is defined as sub-region 1; the cylindrical part below the cylinder hoop is defined as sub-region 2; and the bottom surface of the cylinder is defined as sub-region 3. For sub-region 3, there is no need for further division of its area, nor is there a need for circumferential expansion; therefore, steps S12-S19 of the horizontal salt mixer embodiment can be directly followed.

[0184] The specific description of the deformation detection process for sub-region 2, the cylindrical portion below the bottle neck, is as follows:

[0185] S1. Divide the sub-region 2 below the small gas cylinder hoop into 3 equal parts along the circumference;

[0186] S2. Divide the sub-region 2 below the small gas cylinder hoop into 3 equal parts along the height direction;

[0187] S3. After division, the surface of the gas buffer tank to be tested is divided into 9 sub-regions;

[0188] S4. In each sub-region, a marker point is placed at a distance of 0.1m along the circumferential direction for splicing identification points;

[0189] S5. A marker point is placed at 0.1m intervals for the splicing identification points in the axial direction of each sub-region;

[0190] S6. Using striped structured light at a certain distance from the surface of the gas buffer tank, scan sequentially from sub-region (M1, N1) to sub-region (M3, N3).

[0191] S7. Obtain the point cloud data set of the entire circumference surface of the cylinder in region 2;

[0192] S8. The point cloud data of sub-regions (M1, N1) to (M3, N3) are clipped sequentially, and the point cloud data of the sub-regions outside the X and Y directions are clipped.

[0193] S9. Let Δd be the deformation of the sub-region at a distance perpendicular to the central axis O. Optimize the farthest point cloud data in the Z direction for each sub-region:

[0194] S9.1. Take K = 10, S = 1;

[0195] S9.2, if

[0196] S9.3, Remove point cloud data within the radius range of Δd / 10;

[0197] S9.4 The remaining point cloud data is the optimized point cloud data set;

[0198] S10. The cropped and optimized point cloud data of the 9 sub-regions are stitched together, with a cumulative stitching error of 0.1mm;

[0199] S11. Unfold the three-dimensional gas cylinder model along the circumference to form a rectangle;

[0200] S12. Take the average value of the Z-direction coordinates of all point cloud data in the rectangle, and denote it as the base plane of the unfolded rectangle;

[0201] S13. Denote the portions above and below the base plane as convex deformation and concave deformation, respectively. Start traversing from the point cloud data furthest from the base plane until the Z-coordinate of the traversed point cloud data satisfies: Z i -Z O If the value is ≤0.01, then stop iterating.

[0202] S14. Record the shape of the projected outline enclosed by the point cloud where the traversal stops. This shape is the range in which the deformation occurs.

[0203] S15. Unlike any of the above embodiments, this embodiment uses a method of fitting rectangles, circles and triangles together to fit the shape of the projected outline. The area of ​​the image obtained after fitting is the projected size where the deformation occurs.

[0204] S16. Based on the shape contour of the deformed region, mark its centroid position;

[0205] S17. The sub-region where the center of mass is located is the location where deformation occurs.

[0206] S18. Based on the unfolded rectangular 3D point cloud surface, select the deformation contour range of the concave area;

[0207] S19. Select the point cloud that is farthest from the base plane within the area enclosed by the deformation profile, and read its Z-direction distance value from the base plane. imax This refers to the depth at which the concave deformation occurs.

[0208] In summary, this invention provides a method for detecting container deformation, which has the following advantages compared with the prior art:

[0209] 1) To address the problem of accurately identifying and locating deformed parts, especially small deformed parts, of containers in the petrochemical and refining fields, this invention uses a method of first dividing the area and then dynamically splicing it together. This method first identifies the deformation location within each sub-region, and then, through dynamic splicing of the sub-regions, identifies the deformed part on the entire container area.

[0210] 2) To address the difficulty in accurately quantifying the deformation of containers in the petrochemical and refining industries, this invention proposes a three-dimensional measurement method using laser structured light for measuring deformation on container surfaces. This method quantifies the degree of deformation from two perspectives: the projected area parameter of the deformation region and the deformation depth parameter.

[0211] It should be understood that the embodiments disclosed herein are not limited to the specific structures, processing steps, or materials disclosed herein, but should be extended to equivalent substitutions of these features as understood by those skilled in the art. It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting.

[0212] In the description of this invention, unless otherwise stated, "a plurality of" means two or more; the terms "upper," "lower," "left," "right," "inner," "outer," "front end," "rear end," "head," "tail," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0213] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "connected" and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0214] Certain terms are used throughout this application to refer to specific system components. As those skilled in the art will recognize, the same components may often be referred to by different names, and therefore this application is not intended to distinguish those components that differ only in name and not in function. In this application, the terms “comprise,” “include,” and “have” are used in an open-ended manner and should therefore be interpreted as meaning “including, but not limited to…”. Furthermore, the terms “substantially,” “materially,” or “approximately” as used herein refer to industry-accepted tolerances for the corresponding terms. The term “coupling,” as may be used herein, includes direct coupling and indirect coupling via additional components, elements, circuits, or modules, wherein, for indirect coupling, the intermediate component, element, circuit, or module does not alter the information of the signal but may adjust its current level, voltage level, and / or power level. Inferred coupling (e.g., one element is inferredly coupled to another element) includes direct and indirect coupling between two elements in the same manner as “coupling.”

[0215] The phrase "an embodiment" or "an embodiment" used in this specification means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. Therefore, the phrase "an embodiment" or "an embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment.

[0216] The embodiments of the present invention are given for illustrative and descriptive purposes only, and are not intended to be exhaustive or to limit the invention to the forms disclosed. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of the invention, and to enable those skilled in the art to understand the invention and to design various embodiments with various modifications suitable for a particular purpose.

[0217] While the embodiments disclosed in this invention are as described above, the content is merely for the purpose of facilitating understanding of the invention and is not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and variations in form and detail of the implementation without departing from the spirit and scope disclosed herein; however, the scope of patent protection for this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A method for detecting container deformation, characterized in that, The method includes: S1. Based on the structural characteristics of the container under test, the container's entire circumference is divided to obtain the sub-region division results of the container under test; S2. Based on the sub-region division results, perform three-dimensional laser scanning on each sub-region in sequence to obtain a three-dimensional point cloud model of the container to be tested; S3. For the three-dimensional point cloud model, the projected area of ​​the deformation region is fitted and the depression depth is measured to obtain the projected area parameters and deformation depth parameters of the deformation region. In step S3, the projected area of ​​the deformation region is fitted through the following steps: When the three-dimensional point cloud model is unfolded circumferentially, the longer side of the unfolded rectangle is the perimeter of the container, marked as the X-axis, and the shorter side of the unfolded rectangle is the height of the container, marked as the Y-axis. Calculate the average value of all point cloud data in the unfolded rectangle along the Z-axis, and mark it as the base plane of the unfolded rectangle; The portions above and below the base plane are respectively denoted as convex deformation and concave deformation. The process begins by traversing the point cloud data farthest from the base plane, continuing until the Z-coordinate value of the traversed point cloud data satisfies: Z... i -Z O If Z ≤ ∈, then stop traversing, where Z i Z is the Z-coordinate value of the i-th point cloud data. O is the Z-coordinate value of the point cloud data at the base plane, and ∈ is the deformation threshold; Record the projected contour enclosed by the point cloud where the traversal stops, then the projected contour is the range where the deformation occurs; The projected contour is fitted using geometric shapes, and the area of ​​the deformed contour image obtained after fitting is the projected size of the deformed region.

2. The method for detecting container deformation as described in claim 1, characterized in that, Step S1 includes: Obtain the original structural data of the container under test to obtain the structural features of the container under test; Let the diameter of the container to be measured be D. Divide the container into M equal parts along the circumference, satisfying the following condition: Where m is the circumferential direction division coefficient; Let the height of the container to be measured be H. Divide the container into N equal parts along the height direction, satisfying the following condition: Where n is the height direction division coefficient; After dividing the area in the circumferential direction and the height direction, the surface of the container under test is divided into Q sub-regions, where Q = M × N; Sub-regions at the same horizontal level are numbered in a clockwise order along the circumference, and sub-regions at the same vertical line are numbered in a bottom-up order along the height, resulting in the sub-region division of the container under test. Each sub-region on the circumferential surface of the container is numbered (M...). i N k ), i = 1 to M, k = 1 to N.

3. The method for detecting container deformation as described in claim 2, characterized in that, Step S1 further includes setting splicing recognition points at the boundaries of each sub-region based on the sub-region division results, which includes: Along the circumference of each sub-region, a splicing identification point is set at intervals of a first distance, wherein the first distance is... q is the coefficient of the marker point; Along the height direction of each sub-region, a stitching recognition point is set at every second distance, where the second distance... When setting up splicing recognition points, only one set of splicing recognition points needs to be set at the boundary of every two adjacent sub-regions.

4. The method for detecting container deformation as described in claim 3, characterized in that, Step S2 includes: Based on the sub-region division results, three-dimensional laser scanning is performed on each sub-region in sequence to obtain the sub-region point cloud data set of the container under test; Based on the sub-region point cloud data set and combined with the stitching recognition points, the container under test is dynamically stitched around its entire circumference to obtain the three-dimensional point cloud model of the container under test.

5. The method for detecting container deformation as described in claim 4, characterized in that, The point cloud data set of the sub-region of the container under test is obtained through the following steps: a. For the sub-region to be scanned, set up the scanning device and scan the sub-region in a scanning order from bottom to top and from left to right; b. Adjust the relative angle between the scanning field of view and the surface of the sub-region to be scanned, and return to step a after each adjustment; c. After the scan is completed, the point cloud data set of the sub-region of the container to be tested is obtained.

6. The method for detecting container deformation as described in claim 4, characterized in that, Perform full-circle dynamic stitching using the following steps: Based on the sub-region point cloud data set, the point cloud data of each sub-region is cropped, and the point cloud data outside each sub-region is reduced to obtain the cropping result data of each sub-region. For each sub-region, the cropping result data is optimized using point cloud data to obtain the optimized result data for each sub-region; Based on the optimization results data and combined with the stitching recognition points, the point cloud data between sub-regions is dynamically stitched together to obtain the three-dimensional point cloud model of the entire circumference of the container under test.

7. The method for detecting container deformation as described in claim 6, characterized in that, The cropping result data for each sub-region is obtained through the following steps: Let Q be the set of point cloud data corresponding to the current sub-region. all ; Using the stitched recognition points corresponding to the current sub-region as the boundary, mark the point cloud data set outside the boundary as Q. out The point cloud data set within the boundary is Q. in ; Perform point cloud dataset computation Q on the current sub-region out =Q all -Q in The point cloud data outside the current sub-region is clipped to obtain the clipping result data corresponding to the current sub-region.

8. The method for detecting container deformation as described in claim 6, characterized in that, The optimized results data for each sub-region are obtained through the following steps: Let O be the central axis of the container to be measured. Then the deformation of the container to be measured perpendicular to the central axis O is Δd, where Δd is the height of the protrusion or the depth of the depression of the container to be measured. For the case of convex deformation, the point cloud at the farthest vertical distance from the central axis O is denoted as [the distance from the central axis O is denoted as ]. Let R1 be the number of point clouds at the convex tip. If R1 satisfies the formula: S is the threshold for the number of point clouds, then it is determined that... Within the scope, with Data within the radius is considered to be randomly occurring point cloud data and is deleted; K is the radius coefficient. For concave deformation, the point cloud at the nearest vertical distance from the central axis O has a distance of [missing information]. Let R2 be the number of point clouds at the apex of the depression. If R2 satisfies the formula: Then it is determined Within the scope, with Data within the radius is considered to be accidental point cloud data and is deleted.

9. The method for detecting container deformation as described in claim 6, characterized in that, The three-dimensional point cloud model of the entire circumference of the container under test is obtained by stitching together the following steps: Sub-region (M) i N k ) and sub-region (M) i+1 N k The splicing error at each circumferential splicing recognition point between ) is Δe i Then the cumulative error between the two regions in the circumferential direction is Sub-region (M) i N k ) and sub-region (M) i N k+1 The stitching error at each height stitching recognition point between the two points is Δe. k Then the cumulative error between the two regions in the height direction is Compute sub-region (M) i N k The cumulative error of the splicing recognition points is like δ is the stitching error threshold, then the sub-region (M) is considered to be... i N k Successfully spliced ​​with adjacent sub-regions; Subregion (M) i N k After successful stitching, move to the next stitching sub-region (M). i+1 N k ) or (M i N k+1 ), determine sub-region (M) i+1 N k ) or (M i N k+1 Whether it is successfully stitched with the adjacent sub-regions, and so on, until all sub-regions are successfully stitched together, to obtain the three-dimensional point cloud model.

10. A method for detecting container deformation as described in any one of claims 1-9, characterized in that, Step S3 further includes: identifying the location of the deformation occurrence region based on the projected area parameter of the deformation region, which includes: The shape contour of the deformed region is determined by the projected area parameter, and the centroid position of the shape contour of the deformed region is marked. The sub-region where the centroid is located is marked as the location where the deformation occurs.

11. The method for detecting container deformation as described in claim 1, characterized in that, In step S3, the depth of the depression in the deformation area is measured through the following steps: Select the deformation contour range of the concave area in the deformation contour image; Determine the point cloud that is farthest from the base plane within the deformation profile range of the depression, and read the distance Z of this point cloud from the base plane in the Z-axis direction. imax Record Z imax The depth of the depression in the deformation area.

12. A storage medium, characterized in that, It includes a series of instructions for performing the method steps as described in any one of claims 1-11.

13. A device for detecting the deformation of a container, characterized in that, The apparatus for performing the method as described in any one of claims 1-11 comprises: The sub-region division module is used to divide the container around the test container according to the structural characteristics of the container, and obtain the sub-region division result of the container. The three-dimensional point cloud model module is used to perform three-dimensional laser scanning on each sub-region in sequence based on the sub-region division results to obtain the three-dimensional point cloud model of the container to be tested. The module for projected area parameters and deformation depth parameters is used to fit the projected area of ​​the deformation region and measure the depression depth for the three-dimensional point cloud model, so as to obtain the projected area parameters and deformation depth parameters of the deformation region.

Images