A 3D vision-based annular area reinforcement mesh hole positioning test device and method

Abstract

The present application belongs to the technical field of construction engineering equipment, and specifically relates to a ring area steel mesh hole positioning test device and method based on 3D vision. Mainly includes the following contents: the 3D sensor is driven to scan the steel mesh through a rotating mechanism, and a steel mesh point cloud image is obtained; the steel mesh image is subjected to binaryzation processing, facilitating subsequent image analysis; the steel mesh image is skeletonized by using a 3D refinement algorithm, and a steel mesh skeleton is extracted; the neighborhood integral algorithm is used to find a steel intersection point set; the least square sphere fitting algorithm is used to fit and position a steel binding point; and the Varignon parallelogram method is used to position a steel mesh center hole. The method reduces the consumption of manpower, improves the construction efficiency, and helps to realize the automation and intelligentization of steel mesh area concrete construction.

Description

Technical Field

[0001] This invention belongs to the field of building engineering equipment technology, specifically relating to a test device and method for locating holes in a ring-shaped steel mesh based on 3D vision. Background Technology

[0002] In construction engineering, the positioning of reinforcing mesh not only ensures that the reinforcing bars are in their designed positions but also optimizes subsequent pouring and vibration processes. Proper positioning prevents the mesh from shifting during pouring, ensuring that the concrete is evenly distributed around the reinforcing bars and avoiding honeycomb-like structures or voids. Simultaneously, this facilitates precise vibration and adequate air release during compaction, resulting in denser concrete and ensuring the strength and durability of the structure.

[0003] The positioning of reinforcing mesh is crucial for construction quality and structural safety. Traditional methods rely heavily on manual visual positioning and marking, which is not only inefficient but also prone to inaccuracies due to human error. Especially in complex construction environments, accurately and efficiently positioning the center holes of the reinforcing mesh is a pressing issue that needs to be addressed. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a 3D vision-based experimental device and method for locating holes in annular steel mesh. This invention enables precise positioning of the central holes in the steel mesh and subsequent construction operations. It accurately identifies the skeleton structure and hole locations of the steel mesh, achieving high positioning accuracy while reducing manual intervention and improving construction efficiency. It is applicable to annular steel meshes of different specifications and shapes. This invention significantly improves the recognition stability of complex spatial structures, overcoming the industry challenges of low efficiency in manual inspection and large positioning deviations in two-dimensional vision.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A method for locating holes in a ring-shaped steel mesh based on 3D vision, the method comprising the following steps:

[0007] Step 1: Install the rotating mechanism and fix the 3D camera under the rotating arm of the rotating mechanism;

[0008] Step 2: Start the 3D camera and servo motor. The rotating mechanism, together with the 3D sensor, scans the steel mesh to collect the 3D point cloud data and pixel value distribution information of the steel mesh.

[0009] Step 3: Transmit the acquired image data to the processing unit for binarization and skeletalization to extract the skeleton structure of the steel mesh;

[0010] Step 4: Identify the intersections of the steel mesh using a neighborhood statistical algorithm, and use the least squares method to fit the intersection set to determine the coordinates of the binding points;

[0011] Step 5: Connect adjacent binding points, calculate the geometric center of the quadrilateral using the Warringon parallelogram method, and obtain the position of the center hole of the steel mesh in the annular area;

[0012] Step 6: Feed back the center hole location information to the construction equipment to complete the positioning of the center hole of the steel mesh.

[0013] Furthermore, the steel mesh images acquired by the 3D camera are binarized for subsequent image analysis. A 3D thinning algorithm is used to skeletonize the steel mesh images. By constructing an octree data structure and iteratively thinning the image, the skeleton structure of the steel mesh is extracted. Using any point in the skeletonized image as the center of a sphere, a search is performed with a certain radius. The set of steel mesh intersection points is determined by a neighborhood statistical algorithm. The least squares method is used to fit and locate the binding points. A least squares circle fitting algorithm is used to fit a sphere with each set of points as its edge. The center of the fitted sphere is determined as the binding point. After the binding points are mapped to a two-dimensional plane, the position of the central hole of the steel mesh in the annular region is located by the Warringon parallelogram method.

[0014] Furthermore, the binarization process involves setting the reinforcement value to 1 and the background value to 0.

[0015] Furthermore, the 3D thinning algorithm includes the following steps:

[0016] Initialization: Load the 3D binary image into memory;

[0017] Constructing an octree: Use a 3×3×3 octree data structure to examine the 26 neighborhoods of each pixel;

[0018] Iterative refinement: Identify and remove boundary pixels, generate a candidate list to be removed, and sequentially re-examine the pixels in the candidate list to maintain image connectivity;

[0019] Termination condition: Stop iterating when the image no longer changes;

[0020] Post-processing: Eliminate noise stray points to obtain a smoother, more centered skeleton.

[0021] Furthermore, the neighborhood statistical calculation method is as follows: taking a point in the skeletalized image as the center of a sphere, searching within a certain radius, counting the number of pixels with a pixel value of 1 found within this range, and if the counted number of points is greater than a certain threshold, it is determined to be an intersection point.

[0022] Furthermore, the least squares method for fitting and locating the binding point involves: using a least squares sphere fitting algorithm to fit a sphere with each set of intersection points as its edges, and finally determining the center of the fitted sphere as the binding point; wherein, the standard equation of the sphere is:

[0023] (xa) 2 +(yb) 2 +(zc) 2 =r 2 ;

[0024] In the formula, (x, y, z) is a point on the edge of the sphere, (a, b, c) are the coordinates of the center of the sphere, and r is the radius of the sphere; expanding the equation and rearranging the terms, we can obtain:

[0025] x 2 +y 2 +z 2 -2ax-2by-2cz+(a 2 +b 2 +c 2 -r 2 ) = 0

[0026] Define a new variable d = a 2 +b 2 +c 2 -r 2 Then the above formula can be simplified to:

[0027] -2ax-2by-2cz+d=-(x 2 +y 2 +z 2 )

[0028] Further transformed into standard linear form:

[0029] Ax + By + Cz + D = E

[0030] Where A = -2a, B = -2b, C = -2c, D = d, E = -(x 2 +y 2 +z 2 );

[0031] For a given set of n points (x i ,y i ,z i We can write n such equations, forming a matrix:

[0032] M·p=E;

[0033] in:

[0034] M is an n×4 matrix, in the form of:

[0035] p = [A, B, C, D] T It is the parameter vector to be determined;

[0036] E is an n×1 vector, in the form:

[0037] E = [-(x1) 2 +y1 2 +z1 2 ),-(x2 2 +y2 2 +z2 2 ),...,-(x n 2 +y n 2 +z n 2 )] T

[0038] Solving the equation M·p=E using the least squares method, we obtain p=[A,B,V,D] T ;

[0039] The formula can be solved using matrices:

[0040] p = (M T M) -1 M T E

[0041] Given A = -2a, B = -2b, and C = -2c, we can obtain the coordinates of the sphere's center, which are also the coordinates of the binding point:

[0042] a=-(A) / 2, b=-(B) / 2, c=-(C) / 2.

[0043] Furthermore, the Valenc parallelogram method for locating the reinforcing mesh holes involves calculating the midpoints of adjacent binding points, and the intersection of the midpoints of the lines connecting them is the location of the center hole of the reinforcing mesh in the annular area.

[0044] The present invention also claims a test device for a method of locating holes in a ring-shaped steel mesh based on the above-mentioned 3D vision, comprising a rotating mechanism and a 3D sensor. The rotating mechanism includes a rotating base, a pinion, a gear, a bearing, a rotating rod, and a servo motor. The rotating base is fixed on a support platform and is engaged with the rotating rod through the bearing. The servo motor is fixed on the rotating rod, and the servo motor drives the pinion to cooperate with the gear on the rotating base to drive the rotating rod to move circumferentially.

[0045] Furthermore: a second servo motor is installed at the end of the rotating rod, and a 3D sensor is fixed on the moving block of the ball screw. The second servo motor drives the moving block and the 3D sensor to move radially along the direction of the ball screw.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] (1) In this invention, the binding points are mapped to a two-dimensional plane. The midpoints of adjacent binding points are taken, and a straight line passing through the midpoints of the corresponding points is drawn. The intersection of the straight lines is the location of the center hole of the steel mesh in the annular area. The center hole location information is fed back to the construction equipment through the control unit, so as to realize the precise positioning of the center hole of the steel mesh and subsequent construction operations.

[0048] (2) This invention can accurately identify the skeleton structure and hole location of the steel mesh by using a high-precision 3D camera and image processing algorithm, with high positioning accuracy; combined with a rotating mechanism and automated control technology, it can achieve comprehensive scanning and positioning of the steel mesh in the annular area, reduce manual intervention and improve construction efficiency; it is applicable to steel mesh in the annular area of ​​different specifications and shapes, and has broad engineering application prospects.

[0049] (3) This invention achieves high-precision, adaptive hole positioning through three-dimensional thinning algorithm, least squares sphere fitting and Warrington parallelogram geometry calculation. It also effectively filters noise and accurately extracts steel bar intersections by using skeletal processing and neighborhood statistics. Combined with the dynamic scanning of the rotating mechanism to generate dense point cloud, it significantly improves the recognition stability of complex spatial structures. At the same time, the multi-degree-of-freedom scanning mechanism driven by dual servo motors breaks through the traditional perspective limitation. The gear set and ball screw work together to achieve omnidirectional coverage and stable motion, ensuring the integrity of data acquisition. The two work together to build a closed-loop feedback system. Through three-dimensional spatial coupling optimization algorithm efficiency and fault tolerance, it overcomes the industry problems of low manual detection efficiency and large two-dimensional visual positioning deviation. Attached Figure Description

[0050] Figure 1 This is a three-dimensional schematic diagram of the overall structure of the experimental device of the present invention;

[0051] Figure 2 This is a top view of the experimental apparatus of the present invention;

[0052] Figure 3 This is a three-dimensional schematic diagram showing the disassembled rotating mechanism of the present invention;

[0053] Figure 4 This is a flowchart of the method for locating holes in the reinforcing steel mesh according to the present invention.

[0054] The attached figures are labeled as follows:

[0055] 1. Rotating mechanism; 2. 3D sensor; 1-1. Rotating base; 1-2. Small gear; 1-3. Large gear; 1-4. Bearing; 1-5. Rotating rod; 1-6. Servo motor; 1-8. Servo motor II. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. Of course, the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0057] Although the steps in this invention are arranged by reference numerals, this is not intended to limit the order of the steps. Unless the order of the steps is explicitly stated or the execution of a step requires other steps as a basis, the relative order of the steps can be adjusted. It is understood that the term "and / or" as used herein refers to and covers any and all possible combinations of one or more of the associated listed items.

[0058] Example

[0059] like Figures 1 to 4 As shown, the following steps are required when using it:

[0060] 1) Install a rotating mechanism in the construction area and fix the 3D camera under the rotating rod of the rotating mechanism.

[0061] 2) Start the 3D camera and servo motor, and the rotating mechanism works with the 3D sensor to scan the steel mesh and collect the three-dimensional point cloud data and pixel value distribution information of the steel mesh.

[0062] 3) The acquired image data is transmitted to the processing unit for binarization and skeletonization to extract the skeleton structure of the steel mesh.

[0063] 4) The intersections of the steel mesh are identified by the neighborhood statistical algorithm, and the least squares method is used to fit the intersection set to determine the coordinates of the binding points.

[0064] 5) Connect adjacent binding points, calculate the geometric center of the quadrilateral using the Warinion parallelogram method, and obtain the position of the center hole of the steel mesh in the annular area.

[0065] 6) Feed back the center hole location information to the construction equipment to complete the positioning of the center hole of the steel mesh.

[0066] The 3D vision-based method for locating holes in a ring-shaped steel mesh in this invention can quickly and accurately locate the central holes of the steel mesh in the ring-shaped area, improve construction efficiency and quality, reduce manpower consumption, and has significant engineering application value.

[0067] This invention achieves high-precision, adaptive hole localization through a three-dimensional thinning algorithm, least-squares sphere fitting, and Warrington parallelogram geometry calculation. It also effectively filters out noise and accurately extracts steel bar intersections by using skeletalization processing and neighborhood statistics. Combined with a rotating mechanism to dynamically scan and generate a dense point cloud, it significantly improves the recognition stability of complex spatial structures. This will be described in detail later.

[0068] The experimental device for locating rebar mesh holes in an annular area based on 3D vision includes a rotating mechanism 1, a 3D sensor 2, and a rebar mesh hole positioning method. Its key feature is that the 3D sensor 2 is installed below the rotating mechanism 1, and the rebar mesh hole positioning method locates the rebar mesh holes based on a three-dimensional rebar mesh model.

[0069] Furthermore, the rotating mechanism 1 includes a rotating base 1-1, a pinion 1-2, a large gear 1-3, a bearing 1-4, a rotating rod 1-5, and a servo motor 1-6. The rotating base 1-1 is fixed to the support platform and engages with the rotating rod 1-5 via the bearing 1-4. The servo motor 1-6 is fixed to the rotating rod 1-5 and drives the pinion 1-2 to rotate in conjunction with the large gear 1-3 on the rotating base 1-1, thereby causing the rotating rod 1-5 to move circumferentially.

[0070] Furthermore: 3D sensor 2 is installed below rotating rods 1-5, and its data interface and lens meet the current mainstream models, ensuring that the monitoring range effectively covers the steel mesh.

[0071] Furthermore: a servo motor 1-8 is installed at the end of the rotating rod 1-5, and the 3D sensor 2 is fixed on the moving block of the ball screw 2-7. The servo motor 1-8 drives the moving block and the 3D sensor 2 to move radially along the direction of the ball screw 2-7.

[0072] This invention breaks through the limitations of traditional viewing angles by using a multi-degree-of-freedom scanning mechanism driven by dual servo motors. It achieves omnidirectional coverage and stable motion through the collaboration of gear sets and ball screws, ensuring the integrity of data acquisition. The two work together to construct a closed-loop feedback system, and through three-dimensional spatial coupling optimization algorithm efficiency and fault tolerance, it overcomes the industry problems of low efficiency of manual inspection and large deviation of two-dimensional visual positioning.

[0073] Furthermore, the rebar mesh hole location method includes binarizing the rebar mesh image acquired by the 3D camera to facilitate subsequent image analysis; using a 3D thinning algorithm to skeletonize the rebar mesh image, extracting the skeleton structure of the rebar mesh by constructing an octree data structure and iterative thinning process; using any point in the skeletonized image as the center of a sphere, searching with a certain radius, and determining the set of rebar intersection points through a neighborhood statistical algorithm; using the least squares method to fit and locate the binding points, fitting a sphere with each set of points as its edge through the least squares circle fitting algorithm, and determining the center of the fitted sphere as the binding point; after mapping the binding point to a two-dimensional plane, locating the central hole position of the rebar mesh in the annular region using the Warrington parallelogram method.

[0074] Furthermore, the binarization process involves setting the reinforcement value to 1 and the background value to 0.

[0075] Furthermore, the 3D thinning algorithm includes the following steps:

[0076] Initialization: Load the 3D binary image into memory;

[0077] Constructing an octree: Use a 3×3×3 octree data structure to examine the 26 neighborhoods of each pixel;

[0078] Iterative refinement: Identify and remove boundary pixels, generate a candidate list to be removed, and sequentially re-examine the pixels in the candidate list to maintain image connectivity;

[0079] Termination condition: Stop iterating when the image no longer changes;

[0080] Post-processing: Eliminate noise stray points to obtain a smoother, more centered skeleton.

[0081] Furthermore, the neighborhood statistical calculation method is as follows:

[0082] Using a point in the skeletalized image as the center, a search is performed with a certain radius. The number of pixels with a value of 1 found within this range is counted. If the counted number of points is greater than a certain threshold, it is determined to be an intersection point.

[0083] Furthermore, the least squares method for fitting and locating the binding point is as follows: the least squares sphere fitting algorithm is used to fit a sphere with each set of intersection points as its edge, and finally the center of the fitted sphere is determined as the binding point.

[0084] The standard equation of a sphere is:

[0085] (xa) 2 +(yb) 2 +(zc) 2 =r 2

[0086] In the equation, (x, y, z) is a point on the edge of the sphere, (a, b, c) are the coordinates of the center of the sphere, and r is the radius of the sphere. Expanding the equation and rearranging the terms, we get:

[0087] x 2 +y 2 +z 2 -2ax-2by-2cz+(a 2 +b 2 +c 2 -r 2 ) = 0

[0088] Define a new variable d = a 2 +b 2 +c 2 -r 2 Then the above formula can be simplified to:

[0089] -2ax-2by-2cz+d=-(x 2 +y 2 +z 2 )

[0090] Further transformed into standard linear form:

[0091] Ax + By + Cz + D = E

[0092] Where A = -2a, B = -2b, C = -2c, D = d, E = -(x 2 +y 2 +z 2 ).

[0093] For a given set of n points (x i ,y i ,z i We can write n such equations, forming a matrix:

[0094] M·p=E;

[0095] in:

[0096] M is an n×4 matrix, in the form of:

[0097] p = [A, B, C, D] T It is the parameter vector to be determined.

[0098] E is an n×1 vector, in the form:

[0099] E = [-(x1) 2 +y1 2 +z1 2 ),-(x2 2 +y2 2 +z2 2 ),...,-(x n 2 +y n 2 +z n 2 )] T

[0100] Solving the equation M·p=E using the least squares method, we obtain p=[A,B,V,D] T .

[0101] The formula can be solved using matrices:

[0102] p = (M T M) -1 M T E

[0103] Given A = -2a, B = -2b, and C = -2c, we can obtain the coordinates of the sphere's center, which are also the coordinates of the binding point:

[0104] a=-(A) / 2, b=-(B) / 2, c=-(C) / 2.

[0105] Furthermore, the Valenc parallelogram method for locating the reinforcing mesh holes involves calculating the midpoints of adjacent binding points, and the intersection of the midpoints of the lines connecting them is the location of the center hole of the reinforcing mesh in the annular area.

[0106] This invention maps the binding points to a two-dimensional plane, takes the midpoint of adjacent binding points, and draws a straight line passing through the midpoints of these points. The intersection of these lines is the location of the center hole in the annular reinforcement mesh. The center hole location information is fed back to the construction equipment via a control unit, enabling precise positioning of the center hole in the reinforcement mesh and subsequent construction operations. Through high-precision scanning with a 3D camera and image processing algorithms, the skeletal structure and hole locations of the reinforcement mesh can be accurately identified, resulting in high positioning accuracy. Combined with a rotating mechanism and automated control technology, comprehensive scanning and positioning of the annular reinforcement mesh is achieved, reducing manual intervention and improving construction efficiency. It is applicable to annular reinforcement meshes of different specifications and shapes, and has broad engineering application prospects.

[0107] The above description is only a preferred embodiment of the present invention. It should be noted that those skilled in the art can make several improvements and modifications without departing from the inventive concept of the present invention, and these all fall within the protection scope of the present invention.

Claims

1. A method for locating holes in a ring-shaped steel mesh based on 3D vision, characterized in that, The method includes the following steps: Step 1: Install the rotating mechanism and fix the 3D camera under the rotating arm of the rotating mechanism; Step 2: Start the 3D camera and servo motor. The rotating mechanism, together with the 3D sensor, scans the steel mesh to collect the 3D point cloud data and pixel value distribution information of the steel mesh. Step 3: Transmit the acquired image data to the processing unit for binarization and skeletalization to extract the skeleton structure of the steel mesh; Step 4: Identify the intersections of the steel mesh using a neighborhood statistical algorithm, and use the least squares method to fit the intersection set to determine the coordinates of the binding points; Step 5: Connect adjacent binding points, calculate the geometric center of the quadrilateral using the Warringon parallelogram method, and obtain the position of the center hole of the steel mesh in the annular area; Step 6: Feed back the center hole location information to the construction equipment to complete the positioning of the center hole of the steel mesh.

2. The method for locating holes in a ring-shaped steel mesh based on 3D vision according to claim 1, characterized in that, The steel mesh images captured by the 3D camera are binarized for subsequent image analysis. A 3D thinning algorithm is used to skeletonize the steel mesh images. An octree data structure is constructed, and an iterative thinning process is employed to extract the skeleton structure of the steel mesh. Using any point in the skeletonized image as the center of a sphere, a search is performed with a certain radius. A neighborhood statistical algorithm is used to determine the set of steel mesh intersection points. The least squares method is used to fit and locate the binding points. A least squares circle fitting algorithm is used to fit a sphere with each set of points as its edge, and the center of the fitted sphere is determined as the binding point. After mapping the binding points to a two-dimensional plane, the central hole position of the steel mesh in the annular region is located using the Warrington parallelogram method.

3. The method for locating holes in a ring-shaped steel mesh based on 3D vision according to claim 2, characterized in that, The binarization process involves setting the reinforcement value to 1 and the background value to 0.

4. The method for locating holes in a ring-shaped steel mesh based on 3D vision according to claim 2, characterized in that, The 3D thinning algorithm includes the following steps: Initialization: Load the 3D binary image into memory; Constructing an octree: Use a 3×3×3 octree data structure to examine the 26 neighborhoods of each pixel; Iterative refinement: Identify and remove boundary pixels, generate a candidate list to be removed, and sequentially re-examine the pixels in the candidate list to maintain image connectivity; Termination condition: Stop iterating when the image no longer changes; Post-processing: Eliminate noise stray points to obtain a smoother, more centered skeleton.

5. The method for locating holes in a ring-shaped steel mesh based on 3D vision according to claim 2, characterized in that, The neighborhood statistical calculation method is as follows: taking a point in the skeletalized image as the center of a sphere, searching within a certain radius, counting the number of pixels with a pixel value of 1 found within a certain radius, and if the number of points counted is greater than a certain threshold, it is determined to be an intersection point.

6. The method for locating holes in a ring-shaped steel mesh based on 3D vision according to claim 2, characterized in that, The least squares method for fitting and locating the binding point is as follows: the least squares sphere fitting algorithm is used to fit a sphere with each set of intersection points as its edge, and finally the center of the fitted sphere is determined as the binding point. The standard equation of a sphere is: ； In the formula A point on the edge of the ball. Let the coordinates of the sphere's center be... Let be the radius of the sphere; Expanding the equation and rearranging the terms, we get: Define a new variable Then the above formula simplifies to: Further transformed into standard linear form: in For a given Points Write These equations, in matrix form, are: ； in: yes The matrix is ​​in the form of: ; It is the parameter vector to be determined; It is The vector is of the form: Solving equations using the least squares method ,get ; Solving the formula using matrices: Depend on Obtain the coordinates of the sphere's center, which is also the coordinates of the binding point: 。 7. The method for locating holes in a ring-shaped steel mesh based on 3D vision according to claim 2, characterized in that, The Valenc parallelogram method for locating the reinforcing mesh holes involves calculating the midpoints of adjacent binding points, and then connecting the midpoints of the lines to find the intersection of these midpoints, which is the location of the center hole in the annular area of ​​the reinforcing mesh.

8. An experimental apparatus for locating holes in a ring-shaped steel mesh using 3D vision as described in any one of claims 1 to 7, characterized in that, The device includes a rotating mechanism (1) and a 3D sensor (2). The rotating mechanism (1) includes a rotating base (1-1), a pinion (1-2), a large gear (1-3), a bearing (1-4), a rotating rod (1-5), and a servo motor (1-6). The rotating base (1-1) is fixed on a support platform and is engaged with the rotating rod (1-5) through the bearing (1-4). The servo motor (1-6) is fixed on the rotating rod (1-5). The servo motor (1-6) drives the pinion (1-2) to cooperate with the large gear (1-3) on the rotating base (1-1) to drive the rotating rod (1-5) to move circumferentially.

9. The test apparatus according to claim 8, characterized in that, The end of the rotating rod (1-5) is equipped with a servo motor (1-8), and the 3D sensor (2) is fixed on the moving block of the ball screw. The servo motor (1-8) drives the moving block and the 3D sensor (2) to move radially along the direction of the ball screw.

Images